Effect of supplemental water on morphology, density, survival and population dynamics of Agropyron
smithii and Bouteloua gracilis
by Kurt William Swingle
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in
Biological Sciences
Montana State University
© Copyright by Kurt William Swingle (1986)
Abstract:
Supplemental water was applied to native grassland over four summers and the growth responses of
two clonal grass species were measured. Irrigation was applied at four levels: 1.) Natural rainfall only
(control), 2.& 3.) Natural rainfall plus water supplements equaling a total weekly minimum of 6 mm
and 12 mm water and, 4.) heavy irrigation (a minimum total of 25 mm each week). These irrigation
regimens were implemented on two fields in Eastern Montana, one dominated by Bouteloua gracilis
the other by Agropyron smithii. Culm densities (culms/m 2), measured for Agropyron smithii,
increased with irrigation and declined slowly in the six years after irrigation was discontinued. Within
year survival was recorded for early spring cohorts of Agropyron smithii (20 culms) and Bouteloua
gracilis (30 culms). Agropyron smithii culm survival was slightly enhanced by all levels of irrigation
but Bouteloua gracilis survival was not affected by irrigation. Morphologic characters (culm height,
number of nodes on culms, seasonal length maxima, and total length of green tissue supported by
culms) were also measured.
All of these showed plastic responses to supplemental water; however, only heavy irrigation
consistently produced responses which were significant. Rates of leaf senescence and emergence
(seasonal means among the irrigation treatments) were calculated. No statistical difference among
treatments could be found for these seasonal means. A weak correlation to both plant water potential
and season was found in the leaf emergence rates. Senescence rates were not correlated with season or
water potential. EFFECT OF SUPPLEMENTAL WATER ON MORPHOLOGY,
DENSITY,
SURVIVAL, AND POPULATION DYNAMICS
OF AGROPYRON SMITHII AND BOUTELOUA GRACILIS
by
Kurt William Swingle
A thesis submitted in partial fulfillment
of the requirements for the degree
of
Master of Science
in
' .
Biological Sciences
MONTANA STATE UNIVERSITY
.Bozeman, Montana
November 1986
,VlAlN UB
1)137?
(-'Qf*' ^
0
COPYRIGHT
by
Kurt William Swingle
1986
All Rights Reserved
ii
APPROVAL
of a thesis submitted byKurt William Swingle
This thesis has been read by each member of the
thesis committee and has been found to be satisfactory
regarding content , English usage, format , citations
bibliographic style, and consistency, and is ready for
submission to the College of Graduate Studies.
Date
Chairperson, Graduate Committee
Approved for the Major Department
, /*7 S b
Date
________________________
Head, Major Department
Approved for the College of Graduate Studies
&
Date
^ /*7 tPlC
Graduate Dean
iii
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of
the requirements for a master's degree at Montana State
University , I agree that the Library shall make it
available to borrowers under rules of the Library. Brief
quotations from this thesis are allowable without special
permission, provided that accurate acknowledgement of
source is made.'
Permission for extensive quotation from or
reproduction of this thesis may be granted by my major
professor, or in his absence, by the Director of Libraries
when,
in the opinion of either, the proposed use of the
material is for scholarly purposes. Any copying or use of
the material in this thesis for financial gain shall not
be allowed without my written permission.
Signature
Date "2 4)
Ic^
<*■
iv
TABLE OF CONTENTS
Page
LIST OF
TABLES.........................................
vii
LIST OF
FIGURES........................................
ix
ABSTRACT................................................
x
GENERAL INTRODUCTION...................................
I
The Study Site................. 1..................
Irrigation Regimen..........
Effectiveness of Irrigation Treatments...........
Questions Addressed...............................
PART I.
The Effect of Supplemental Water on Culm
Density in a Stand of Agropyron smithii.....
I
2
4
8
9.
INTRODUCTION..................
10
METHODS............................................
11
RESULTS.................... .
......... ;.........
13
Initial Field Conditions....................
Intraseasonal Dynamics......................
Control Plot...........*......... .
6 mm Plot...............................
12 mm Plot......
Wet Plot................................
Intraseasonal Dynamics......................
1977 vs. 1978..........................
1979 ......................
1980 ...................................
1981', 1982, 1983, & 1986...............
13
13
14
17
18
18
19
20
20
21
22
DISCUSSION..............................'....... . ..
23
Intraseasonal Dynamics..................
Interseasonal Dynamics......................
CONCLUSIONS
23
24
26
V
TABLE OF C O N T E N T S - - Continued
Page
PART TI. The Effect of Supplemental Water on Culm
Mortality In Stands of Agropyron smithii
and Bouteloua gracilis........................
28
INTRODUCTION...............
29
METHODS............................................
RESULTS............................................
30
32
Agropyron
Agropyron
Agropyron
Agropyron
Bouteloua
smithii
smithii
smithii
smithii
gracilis
Control...................
6 m m ..............
12 m m .....................
W e t . . ......................
Control, 6 mm, and Wet...
32
34
34
35
35
DISCUSSION.........................................
37
CONCLUSIONS.......................
39
Agropyron smithii............................
Bouteloua gracilis...........................
39
39
PART III. Effect of Supplemental Water on Leaf
and Culm Morphology In Agropyron smithii
and Bouteloua gracilis.......................
40
INTRODUCTION.......................................
41
METHODS..'..........................................
43
RESULTS AND DISCUSSION............................
46
Culm Height..................
1978 ....................................
1979 .....................................
Culm HeightConclusions.................
Mean Number of Nodes perCulm................
Agropyron smithii......................
Bouteloua gracilis.....................
Mean Number of Leaves perCulm...............
Initial Values..........................
1977 Agropyron smithii.................
1978 Agropyron smithii.................
1979 Agropyron smithii.................
1978 Bouteloua gracilis................
1979 Bouteloua gracilis................
46
46
49
50
50
50
53
54
55
57
57
58
59
59
vi
TABLE QF CONTENTS— Continued
Page
Mean Leaf Number Conclusions................
Maximum Leaf Length- Agropyron smithii.....
Mean Green Tissue per Culm..................
1977 .....................................
1978 ................................... ..
1979 ...............................
Green Tissue Length Conclusions.......
Density Effects..............................
60
60
63
63
65
66
66
67
CONCLUSIONS........................................
68
PART IV. Senescence , Emergence, and Maturation
Rates in Leaves of Agropyron smithii
and Bouteloua gracilis.......................
69
INTRODUCTION.......................................
Scope of the Study...........................
Concepts Related to Leaf Dynamics...........
70
70
71
METHODS............................................
73
RESULTS......
75
Seasonal Averages of
Emergence and Senescence Rates..............
Comparison of Treatments...............
Comparison of Years....................
Difference Between the Two Rates......
Seasonal Influence on Leaf
Emergence and Senescence Rates........... ...
Effect of Water Potential on
Emergence and Senescence Rates..............
Graphical Analysis of Emergence.......
,Graphical Analysis of Senescence......
Statistical Analysis................
Leaf Juvenile Period........................ -.
Leaf Adult Interval..........................
75
75
77
78
79
80
80
85
85
87
88
DISCUSSION. :.......................................
90
CONCLUSIONS........................................
93
REFERENCES CITED....................
94
APPENDICES...............................................
98
Appendix A ....................................
Appendix B ...................................
99
102
vii
LIST OF TABLES
Table
Page
1. Average temperature and precipitation (1941-1970)
for Miles CityFAA (NOAA 1972).....................
2
2. Natural precipitation (cm) at Miles City, MT:
1977-1985 ..........................................
7
3. Large culm density (culms/.03 m 2) during treated
(1977 - 1980) and post-treatment (1981 - 1986)
stand development..................................
21
4. Agropyron smithii culm survivorship as % of
original culms alive at a given date for four
irrigation treatments, 1977 - 1978...............
33
5. Agropyron smithii culm survivorship as % of
original culms alive at a given date for four
irrigation treatments (1979)...................... 34
6. Bouteloua gracilis culm mortality expressed as %
of original culms alive at a given date for
three irrigation treatments.......................
36
7. Mean culm height for Agropyron smithii in
1978 & 1979 ........................................
47
8. Mean culm height for Bouteloua gracilis in
1978 & 1979 ........................................
48
9. Mean nodes per culm in Agropyron smithii during
1977 - 1979 ........................................
52
10. Mean nodes per culm in Bouteloua gracilis during
1978 - 1979 .......................................
54
11. Green leaves/culm in Agropyron smithii during
1977 , 1978 , and 1979 .............................
56
12. Green leaves/culm in Bouteloua gracilis during
1977 (start-of-season only), 1978, and 1979.....
59
13. Mean maximum length (mm) of Agropyron smithii
leaves subjected to four irrigation treatments.. 62
viii
List of T abIes— Continued
Table
,
Page
14. Green leaf tissue length (mm/culm) of Agropyron
smithii 1977-1979................................
64
15. Mean weekly leaf emergence and senescence rates
for Agropyron smithii under four different
irrigation regimens in the summers
of 1977-1979............
76
16. Mean weekly leaf emergence and senescence rates
for Bouteloua gracilis under three different
irrigation regimens for the years
1978 to 1979 ..................................
77
17. Mean monthly leaf emergence rates for Agropyron
smithii (1977-1979) and Bouteloua gracilis
(1978-1979).......................................
79
18. Mean monthly leaf senescence rates for Agropyron
smithii (1977-1979) and Bouteloua gracilis
(1978-1979).......................................
80
19. Mean June leaf emergence and senescence rates
for three levels of water stress in
Agropyron smithii................................
85
20. Mean June leaf emergence and senescence rates
for three levels of water stress in
Bouteloua gracilis.................... -..........
86
21. Mean juvenile period (days) for leaf cohorts of
Agropyron smithii............................. .
87
LIST OF FIGURES
Figure
Page
1. Agropyron smithii water potential 1977 - 1979 .....
5
2. Bouteloua gracilis water potential 19 77 - 1979....
6
3. Agropyron smithii culm
density 1977 ...............
15
Agropyron smithii culm density 1978...............
16
5. Agropyron smithii leaf emergence rate
vs. water potential...............................
81
6. Bouteloua gracilis leaf emergence rate
vs. water potential...................
82
7. Agropyron smithii leaf senescence rate
vs. water potential........... :...................
83
8. Bouteloua gracilis leaf senescence rate
vs. water potential...............................
84
4.
X
ABSTRACT
Supplemental water was applied to native grassland
over four summers and the growth responses of two clonal
grass species were measured. Irrigation was applied at
four levels: I.) Natural rainfall only (control), 2.& 3.)
Natural rainfall plus water supplements equaling a total
weekly minimum of 6 mm and 12 mm water an d , 4.) heavy
irrigation (a minimum total of 25 mm each week). These„
irrigation regimens were implemented on two fields in
Eastern Montana, one dominated by Bouteloua gracilis the
other by Agropyron smithii. Culm densities (culms/m *),
measured for Agropyron smithii, increased with irrigation
and declined slowly in the six years after irrigation was
discontinued. Within year survival was recorded for early
spring cohorts of Agropyron smithii (20 culms) and
Bouteloua gracilis (30 culms). Agropyron smithii culm
survival was slightly enhanced by all levels of irrigation
but Bouteloua gracilis survival was not affected by
irrigation. Morphologic characters (culm height, number of
nodes on culms, seasonal length maxima, and total length
of green tissue supported by culms) were also measured.
All of these showed plastic responses to supplemental
water; however, only heavy irrigation consistently
produced responses which were significant. Rates of leaf
senescence and emergence (seasonal means among the
irrigation treatments) were calculated. No statistical
difference among treatments could be found for these
seasonal means. A weak correlation to both plant water
potential and season was found in the leaf emergence
rates. Senescence rates were not correlated with season or
water potential.
I
GENERAL INTRODUCTION
In the semi-arid northern Great Plains, shortage of
water most limits formation of plant biomass (Whittaker 1975
P . 202). The objective of the project described in this
thesis was to measure the effects of irrigation on five
plastic morphological characteristics and on the population
dynamics of two important range grasses, Agropyron smithii
and Bouteloua gracilis.
To measure the effects of increased water, comparisons
were made among plants grown under four different levels of
irrigation. The morphological and population responses and
methods for their measurement are discussed separately in
the four main sections of this thesis. The study site and
the irrigation treatments used are described first.
The Study Site
The study was conducted under conditions thought to be
representative of the Northern Great Plains at the Fort
Keogh U.S.D.A. Livestock and Range Research Station, Miles
City, Montana. The study plots were located on two level,
ungrazed fields. The vegetation on one was a relatively pure
stand of Agropyron smithii while the second was dominated by
Bouteloua gracilis. The sites were chosen for their
topographic homogeneity, monospecific composition, and
2
availability of the water with which treatments were made.
The soil of the Agropyron site was identified by -USDA/SCS
personnel as a Kobar Silty Clay loam and the Bouteloua field
was classified as having a Havre loam (M. Nichols, personal
communication). A more detailed description of the study
sites appears in Weaver e_t a_l. (1981).
Climatic conditions of the area are summarized in Table
I.
Table I. Average temperature and precipitation (1941-1970)
for Miles City FAA (NOAA 1972).
T e m p .(C)
P p t .(cm )
May
13.5
5.2
June
18.3
8.4
July
23.6
3.9
August
22.5
3.0
Annual
7.4
35.4
Irrigation Regimen
Four irrigation treatments were applied on two sites
for four years. Two levels of irrigation (6 mm and 12mm)
were chosen to simulate results of two levels of weather
modification success. Plant responses under these treatments
were compared with plant responses under unwatered
(control),
and heavily watered (wet) conditions. The object
of the two simulated levels was to eliminate all rain-free
periods (droughts) of over a week's duration. The quantities
of water deliverable under weather modification were not
known and are still under investigation (Barge e_t a_l. 1986).
This being the case, two levels of weekly supplemental
3
increase were arbitrarily selected: 6 mm and 12 mm of water.
If natural precipitation was inadequate to meet these
levels,
irrigation was applied until the weekly total met
these minima .
The water treatments were applied with sprinkler
irrigation on four plots at each site. The "control" plot
received only natural precipitation. Measured amounts of
water were supplied to two other plots such that during each
week each plot would receive a minimum of 6 mm or 12 mm of
water. The fourth, "wet", plot was irrigated until soil
moisture blocks (Taylor e_t a_l. 1961) placed at depths of 25
and 75 cm showed a water potential of -0.2 MPa or less. In
the absence of natural precipitation, the amount of
irrigation required to maintain the high water potentials of
the wet treatment was approximately 25 mm/week.
The size of each study plot was 14 x 14 meters. Water
sprinklers were placed 50 cm above the ground and configured
to provide uniform distribution of water over an area 19 x
19 m including the plot. Irrigation was performed only
between the hours of 3:00 A .M . and 9:00 A .M . to minimize
evaporation. Also, to prevent drifting water, irrigation was
restricted to periods when winds were less than 13 cm/sec. A
detailed description of the treatment procedures has been
written by Weaver _e_t a_l. (1981).
4
Field data were collected principally by John Newbauer
(1977-1979),
and in 1980-1981 by Carol Johnson, Brent
Haglundpand Tad Weaver.
Effectiveness of Irrigation Treatments
The degree to which watering treatments were effective
in relieving water stress was documented by measuring plant
and soil water potentials. Whole plant water potentials of
Agropyron smithii and Bouteloua gracilis were measured
throughout the field season with a Scholander pressure bomb
and with the methods described by Ritchie and Hinkley
(1975). Measurements were made at or before dawn on randomly
selected plants.
Results of these pressure bomb measurements are
summarized in. Figures I and 2. When plant water potentials
are lower than -2.0 MPa, plant physiological growth
processes largely stop functioning (Hsiao 1973).
Accordingly,
in these figures,
those measurements which are
are less than -2.3 MPa are classed as infinitely low. The
reason water potential values from different treatments
overlap, as shown in Figures I and 2, is that experimental
design allowed equal amounts of water to fall on all.
treatments when natural rainfall exceeded the treatment
minima.
WATER POTENTIAL, *
(MPa)
5
1977
0.0 q
—0 .5 - \
- 1.0 :
-1 .5 i
-2.0 I
-O O
1978
&
1979
MAY
JUNE
JULY
AUGUST
Figure I. Agropyron smithii water potentials 1977 - 1979.
The four irrigation treatments are: control
(C), 6 mm (6), 12 mm (12), and wet (W). Points
are means of five Scholander pressure bomb
measurements. Significance of differences from
the control on a given date are indicated by
single (p <.05) and double (p <.01) asterisks.
6
0.0
-
-0 .5 1.0
-
WATER POTENTIAL, f
(MPa)
-1 .5
1977
2.0
-
—
2.0
-
1.0
-
1978
-1.5-
1979
MAY
JUNE
JULY
AUGUST
Figure 2. Bouteloua gracilis water potentials 1977 - 1979.
The three irrigation treatments a r e : control
(C), 6 mm (6), and wet (W). Points are means of
five Scholander pressure bomb measurements.
Significance of differences from the control on
a given date are indicated by single (p <.05)
and double (p <.01) asterisks.
7
Throughout this thesis it should be remembered that the
treatment levels were not constant. In the dry summer of
1979, irrigation treatments resulted in very great
differences in water potential among the treatment plots.
During periods of plentiful rainfall (Table 2), no
supplemental water was needed to meet the water minima and
all four treatments received the same amount of (natural)
precipitation. In this situation the control plants received
as much water as the wet treatment plants. This happened
frequently in May and July of 1978, when natural rainfall
was much higher than normal.
Table 2. Natural precipitation (cm) at Miles City, MT:
1977-19851 .
Year
L
Normal
1977
1978
1979
1980
1981
1982
1983
1984
1985
Sept.
April May
14.7
5.2
10.8
6.2
20.5
17.3
24.7
3.5
5.6
0.7
9.0
7.3
14.1
6.6
15.9
3.5
10.7
2.3
8.9
2.9
June
8.4
3.5
3.5
2.0
7.7
6.5
13.0
4.0
9.0
2.4
July
3.9
4.9
6.4
7.1
1.3
0.9
1.8
4.8
0.5
7.9
Aug . Summer
20.7
3.0
5.7
20.3
2.1
29.3
I .7
14.2
5.2
15.0
2.8
17.6
1.5
22.9
0.8
13.1
2.3
14 .I
4.7
17.9
2
Annual
35.4
31 .I
49.8
38.9
26.6
37.0
37.0
29.0
24.8
26.8
Precipitation data from NOAA Climatological Data, Montana,
Miles City FAA (1976 - 1985).
Summer precipitation was calculated as the sum of Ma y ,
June, July, and August rainfall.
Yearly precipitation was calculated by summing monthly
precipitation from September of the previous year though
August of the present year.
Seasonal Normals are averages covering the years 1941 to
1971 .
8
Questions Addressed
The different degrees of release from water stress
resulting from the four irrigation regimens was expected to
product a variety of plastic modifications in both community
and individual plant structure. Four questions regarding
plastic responses were considered in detail:
I.) Would water
supplements change culm density? 2.) Would water supplements
cause an increase in culm survival? 3.) Would increased
water cause morphological changes in these grasses? 4.)
Would water supplements change leaf population dynamics?
These topics are addressed individually in Sections I
through IV.
In general, the plant response to an increase in a
formerly limiting resource (in this case water) is expected
to be more abundant, longer lived, and larger individuals
(Simpson 1981 p. 116, Kramer 1969 pp. 356-360). However as a
consequence of such changes, density dependent factors such
as competition for light or mineral nutrients, may become
increasingly important, and this in turn may affect density,
survival, and size in the opposite direction. For this
reason , both the magnitude and the direction of plastic
response were of interest in this study.
9
PART I
The Effect of Supplemental Water on Culm Density
in a Stand of Agropyron smithii
10
INTRODUCTION
The density of a stand of plants may be regulated by
a variety of factors such as predators, pathogens, and
competition for scarce resources (Antonivics and Levin
1980) . While the physiological effects of water stress ■
(Hsiao 1973, Simpson 1981) and plant density dynamics have
been discussed (Harper 1977 pp. 151-381), no field
experiments have determined the effects of water
availability on plant density. This section describes
increases in culm density of a clonal plant, Agropyron
smithii, as a growth response to irrigation, and the
subsequent decrease in density after irrigation was
discontinued.
11
METHODS
To determine Agropyron smithii culm-density response
to increased water, portions of a pure stand of the grass
were treated with four levels of irrigation. Culms were
counted in permanent quadrats located within the study
plots, to give a record of density fluctuation experienced
by the Agropyron population during irrigation years
(1977-1980) and post-irrigation years (1981 - 1986).
Ten permanently, marked 10 x 30 cm quadrats were
placed at I m
intervals along the central section of an
untrampled 2 x 14 m strip in the center of each study
plot. These quadrats were used throughout the project.
Sampling was done at approximately weekly intervals. The
culms of Agropyron smithii present in these quadrats were
grouped according to size classes. The two classes
consisted of those plants with one. to two leaves, and
those with three or more leaves.
The sampling regimen was maintained from May to late
August of 1977 and 1978. Irrigation treatments were
discontinued at the end of 1980. Isolated density
measurements were made on 16 May and 28 August in 1979; 21
August in 1980; 28 June, 20 July , and 27 August in 1981;
25 July in 1982;
15 June in 1983; and 24 June in 1986.
12
Statistical analysis of the results were made using
Student's t multiple comparisons of means in 10 quadrats
against the control treatment mean. Bartlett's test for
homogeneity of variances showed unequal variances among
the treatments. Efforts at transformation of the data
proved unsatisfactory, so the analysis was made with a
method for comparison of means with unequal variances and
unequal sample sizes, as described by G . W . Snedecor and
W . G . Cochran (Statistical Methods 1980 Chapter 12).
13
RESULTS
Initial Field Conditions
Before irrigation treatments were started,
measurements were made in the test plots to determine the
initial degree of uniformity at the site. On 24 May 1977,
the date of the first measurements, culm densities for >2
leaf culms (which shall be termed "large culms") were
22.2/quadrat ±1.9 S.E. for the control plot, 25.5/quadrat
+2.1 S.E. for the 6 mm plot, and 26.3/quadrat +2.8 S.E.
for the wet plot. Student's t-test did not show heterogeneity in culm density between sites at the 0.05%
confidence level. Multiplication of plot data by 33.33
converts plot data to a I m 2 area. This yields an initial
density of 740 large culms/m 2 in the control plot. Culms
with _<2 leaves (which shall be termed "small culms") had
an initial density of 2.4/quadrat in the control plot, 1.4
in the 6 mm plot and 0.8 in the wet plot.
Intraseasonal Dynamics
Intraseasonal density-dynamics are illustrated in
detail with 1977 - 1978 data, since plant numbers were .
recorded at regular intervals only in these two years.
14
Control Plot. Within a given season, water conditions
varied considerably, and this affected culm numbers. Water
was usually most abundant in the spring, and became less
so through the summer (Figure I). The control plot
experienced intraseasonal fluctuations in culm density,
which paralleled changes in water availability.
During the first part of the 1977 season , control
culm numbers changed little (Figure 3). After 12 July,
however, density of the large culms decreased steadily to
an end-of-season value of 12.9 culms/quadrat or roughly
half the early season value. Relatively rare small culms
also showed little early season variation. After 10 August
the density of small culms increased to a value of 11.8
culms/quadrat.
In 1978 (Figure 4), numbers of large culms in the
control plot increased through the early season. By late
June the density had reached 25.3 culms/quadrat or 843
culms/m2. After that time, large culm density decreased
but only moderately.
Small culms in 1978 decreased
steadily from a high of 9.8 culms/quadrat early in the
season; and reached a density of 0.3 culms/quadrat on I
July. Small culms remained at these relatively low
densities for the remainder of 1978.
15
CULMS /
.03
CULMS WITH > 2 LEAVES
CULMS WITH s 2 LEAVES
MAY
JUNE
JULY
1977
AUGUST
Figure 3. Agropyron smithii culm density 1977. The four
irrigation treatments are: control (C),
6 mm (6), 12 mm (12), and wet (W). Comparisons
of means are made for larger culms only.
Significance of differences from the control on
a given date are indicated by single (p <.05)
and double (p <.01) asterisks.
16
CULMS WITH > 2 LEAVES
CULMS WITH < 2 LEAVES
(Z) 40 -
I
MAY
w
i
o
JUNE
JULY
1978
i
v
r
AUGUST
Figure 4. Agropyron smithii culm density 1978. The four
irrigation treatments are: control (C),
6 mm (6), 12 mm (12), and wet (W). Comparisons
of means are made for larger culms only.
Significance of differences from the control on
a given date are indicated by single (p <. 05)
and double (p <.01) asterisks.
17
6 mm Plot. Dynamics of culm populations in the 6 mm
plot for 1977 parallel those of the control plot (Figure
3). Culm numbers for this year were slightly higher than
those in the control plot except at the beginning of the
season. During the entire 1977 season the density of large
culms deviated only slightly from the initial value of 26
culms/quadrat.
In early July, in the 6 mm treatment, the
density of large culms fell (as it did in the control). By
the end of August large culm densities in the 6 mm plot
had increased to 24.7 culms/quadrat. The density of small
culms in the 6 mm treatment was low and fluctuated little.
In 1977,
small culms increased in density at season's end,
as in the control plot, from a low density of 0.2/quadrat
on 26 July,
to 16.2/quadrat on 30 August.
Seasonal density dynamics for 6 mm culms in 1978 were
undifferentiable from those of the control. Density of
large culms initially increased,
leveled off, then
decreased slightly. Density of small culms decreased from
an early season maximum and stayed at low densities for
the remainder of the season.
Initially,
section),
in 1979 (Table 3 in the following
the 6 mm plot had a large culm density of 8.9
culms/quadrat, not significantly different from the
control. By the end of the field season this density had
increased to 12.9 culms/quadrat.
Small 6 mm treatment
18
culms also increased over time from 0.9 culms/quadrat to
1.7 culms/quadrat.
12 mm Plot. Measurements of culm density in the 12 mm
plot were initiated in 1978 (Figure 4). Seasonal dynamics
for this treatment in 1978 were not significantly
different from either the control or the 6 mm plot. This
was true for both the large and small 12 mm treatment
culms.
In 1979 (Table 3), the 12 mm plot had an early season
density of 26.8 large culms/quadrat. This was
significantly higher than the control. At season's end the
density was essentially unchanged with 17.7 large
culms/quadrat. Small culms increased slightly from
densities of 2.0 culms/quadrat to 7.0 culms/quadrat.
Wet Pl o t . A spectacular increase in culm density
resulted from the wet treatment. In the spring of 1977
(Figure 3), the density of large culms in the wet plot was
similar to that in the control at 26 large culms/quadrat.
This density increased continually through the season and
by 30 August was 60.8 culms/quadrat or 2027 culms/m2.
Small culm numbers did not show the same rapid increase as
did the large culms,
treatments.
but paralleled those of the other
Initial small culm density was 0.8
culms/quadrat which increased only slightly through the
summer. At the very end of the season however the small
19
culms showed an increase similar to that seen in both the
control and 6 mm treatments with an end of season value of
21.7 small culms/quadrat.
In 1978 (Figure 4), culm density in the wet plot
continued to increase for the first quarter of the season,
reaching a maximum density of 83.3 culms/quadrat on 16
May. This is an increase of 625% over the control plot on
the same observation date. After that maximum, culm
density in the wet plot decreased gradually for the rest
of the season except for a slight increase at season's
end. In 1978 small culms were'initially 25.7/quadrat. This
value fell to low densities and remained low for most of
the summer. In August small culm densities increased again
to values similar to the early season. The wet plot was
the only treatment in 1978 to show the late August
increase in small culm density seen in all treatments in
1977 .
The early season 1979 wet plot density was 40.8 large
culms/quadrat and 2.6 small culms/quadrat. End of season
figures were little changed from these values.
Intraseasonal Dynamics
Because small culms represent only a pool of
potential recruitment prospects with unknown overwintering
potential , results in this section consider only.large
culms.
20
1977 vs. 1978. In 1977 the control plot exhibited a
net decrease in culm density (Figure 3). In 1978 the end
of season control plot density (large culms) had increased
both over the course of the summer and in comparison to
the end of season 1977 (Figure 4). The difference in
natural rainfall between the two years and the absence of
water stress for the first half of 1978 can account for
these observations.
In the dry summer of 1977, a small amount of
supplemental water (conditions of the 6 mm treatment)
appeared to inhibit the decrease in culm density
experienced by the control plot. In the 6 mm treatment
plots , there was little change in density over the course
of the year. In 1978, the culm density of the 6 mm plot
increased until some water stress was experienced;
then
it, like the control plot, stopped it's increase but at a
density higher than at season's start.
The wet plot started 1978 with a density similar to
that which it had at the end of season in 1977 (large
culms). While the wet plot showed some early season
increases its net 1978 change was very slight. This
suggests that at the end of two years of treatment
limiting factors other than water may have exerted the
dominant influence on density.
1979. Early in the season of 1979 all treatments,
with the exception of the 12 mm, supported fewer culms
21
than their respective end of season values in 1978 (Table
3). By the end of 1979, as in the previous dry year 1977,
the control density had decreased while the 6 mm treatment
density had changed little. The wet treatment showed no
net increase for 1979. Culms in the 12 mm plot in 1979
also showed no net yearly density increase.
Table 3. Large culm density (culms/.03 m 3) during treated
( 1977 - 1980) and post-treatment ( 1981 - 1986)
stand development. Statistically (Student's t)
significant differences from control means are
indicated. A single asterisk (*) indicates p
<.05, a double asterisk (**) indicates p <.01.
Date
6-21-77
7-26-77
8-30-77
6-21-78
7-18-78
8-30-78
5-16-79
8-28-79
8-21-80
Control
24.8
15.9
12.9
25.3
25.0
20.8
10.7
3.3
2.0
6-28-81
7-20-81
8-27-81
7-25-82
6-15-83
6-24-86
9.1
4.6
0.5
9.0
6.7
16.0
Treatment
6 mm
12 mm
26 .I
21.7
24.7**
25.6
26.8
27 .I
27.4
23.5
22.8
26.8**
8.9
12.9**
27.8**
9.5**
19.3**
Wet
30.4
37.9**
60.8**
70.4**
71.0**
70.7**
81.6**
82.2**
75.2**
15.6*
4.5
0.1
16.2*
11.6*
19.2
89.3**
55.0**
62.5**
64.5**
44.6**
35.8**
—
—
—— —
—
—
—
—
-----—
—
27.4**
6 .I
2.0
21.7**
17.6**
15.2
Irrigated
NonIrrigated
1980. Only one data point exists for 1980, allowing
only an end of season comparison with 1979 (Table 3).
Density in the wet treatment still far exceeded that found
in the other treatments.
22
1981, 1982, 1983, & 1986. Supplemental irrigation was
discontinued at the end of 1980. After this, occasional
measurements were made of culm density to document the
rate of return to equilibrium.
In June 1981 , all irrigated plots had higher culm
densities than the control. For large culms in the wet
plot, end of season culm density was 125 times higher than
that found in the control plot.
In 1982, 1983, and 1986, the effect of previous
irrigation remained detectable. In the wet plot, large
culms for these years were respectively seven, six, and
two times more numerous than the control. Even six years
after supplemental water was no longer being provided,
culm density remained significantly higher in the wet plot
than the control.
J
23
DISCUSSION
Changes in culm density can result from changes in
rates either of culm mortality or of culm emergence. The
time required for a response from either factor to
supplemental water need not be constant over the course of
a season. Root and rhizome density may be altered by the
treatments and these density changes may persist after
termination of the treatment. For these reasons, the
results of the treatments must be evaluated in terms of
intra- and inter- seasonal effects.
Intraseasonal Dynamics
Because each year experienced a different natural
rainfall, during some periods the amount of irrigation
required to provide the required treatment was O mm
whereas at other times it required considerable additions
of water. In order to compare the effect of the treatment
with the control, only contrasts with the control at a
given date are strictly appropriate.
We cannot assume that a fixed percentage of small
culms mature into the class of large culms. It is possible
that a culm may emerge, reach a stage of growth of 2+
leaves and then either die, remain in a stable immature
24
phase or "oscar" (Silvertown 1982, p. 20), or grow into
the large culm class.
From Figure 3 it can be seen that only during the
times when the small culms are above 0 density does the
density of the large culms increase. This suggests that
the number of small culms in a plot is a measure of culm
emergence. This emergence seems to have an interseasonal
dynamic in that small, culms are primarily formed only at
the beginning and end of the season.
Interseasonal Dynamics
Reducing and maintaining water stress at low levels
might have a cumulative effect which was carried over to
the following years. This would influence comparisons made
between years. Even if no moisture were retained in the
soil from the previous summer, the plants might exhibit a
plastic response to water presence that would take several
seasons to disappear.
After the irrigation treatments were discontinued in
1980, culm density continued to be higher in the wet plots
(Table 3). That a growth response persisted has been
corroborated by clip measurements (Weaver 1983) . Possible
reasons for this phenomena are that large amounts of
supplemental water: I .) increased the root depth which
allowed a different horizon of the water table to be
exploited. 2.) increased ground litter which inhibited
25
soil water evaporation. 3.) increased the stand density
past a threshold beyond which plants might be more
successful in excluding invading weedy species.
This aspect of the project deserves further study. In
addition to the possible value for models of successional
processes,
land use managers might benefit from the
knowledge that pasture productivity is increased above
normal levels for years after either a wet seasons' end or
after cessation of heavy irrigation.
26
CONCLUSIONS
Supplemental water generally increases culm density.
Large water supplements (wet treatments) significantly
increased culm density both under relatively wet, and dry,
background climatic conditions. In dry years (1977 &
1979),' small to moderate amounts of supplemental water (6
mm and 12 mm treatments) produced a significant increase
by season's end. In a wet year (1978) small (6 mm
treatments) and moderate (12 mm treatments) amounts of
supplemental water produced no observable density effects.
After three years (two dry and one wet) application of
supplemental water to Agropyron smithii grasslands, the
effect of large (wet) and moderate (12 mm) amounts of
water was a significant increase, in culm density over that
found in control plots. The application of small amounts
of water (6 mm), however, produced no significant changes
in culm density from controls.
New growth of small culms starts frequently at the
end of summer. Water treatments did not appear to
influence this phenomena consistently. This end-of-season
growth took place in all treatments in 1977, but only in
the wet treatment"in 1978.
Small amounts of water had little effect on culm
density. Large amounts of supplemental water had large
27
results. The effects of large amounts of supplemental
water applied over a period of four years, were still
observable six years after water supplements were
discontinued.
PART II
The Effect of Supplemental Water on Culm Mortality In
Stands of Agropyron smithii and Bouteloua gracilis
29
INTRODUCTION
Culms of perennial grass plants are subject to
various risks in the course of a summer. A culm may be
eaten, mechanically damaged, shaded out, or exposed to
physiological stresses. Among physiological dangers, lack
of water is of major importance as a factor limiting
yields of grassland plants (Simpson 1981 pp. 14-16).
If water stress were eliminated or lessened, would
culms persist without appreciable mortality during the
course of a summer? Could small amounts of water (such as
might result from cloud seeding, or small climatic shifts)
ameliorate mortality? Would large amounts of water (such
as heavy irrigation) prove to have an even greater
efficacy? Would the responses of two major grasses of the
Northern Great Plains, Agropyron smithii and Bouteloua
gracilis, to supplemental water be the same? The purpose
of this section is to to answer these questions.
Water stress was artificially relieved in separate
stands of Agropyron smithii and Bouteloua gracilis by
supplementing natural rainfall with measured amounts of
irrigation water. Irrigation treatments were designed to
mimic normal conditions, light periodic showers, and heavy
irrigation. Survival was recorded for individual culms
within each irrigation treatment.
30
METHODS
In order to study culm mortality of Agropyron smithii
and Bouteloua-gracilis plants under varying water
regimens, irrigation water was applied to four separate
plots in two fields each dominated by one of these
grasses. Culms were tagged in late spring and their
survival was recorded during the course of the summer..
Plants- were observed at approximately one week
intervals during the years 1977, 1978, and 1979 for
Agropyron smithii;-1978 and 1979 for Bouteloua gracilis.
Early in.the season, individual culms (20 for Agropyron
smithii and 30 for Bouteloua gracilis) were selected
approximately 75 cm apart within a central untrampled 2 x
14 m strip within the treatment. Culms were selected at
regular intervals along the center of the strip as
representative of those found in the stand. Tags of cotton
thread were tied around their bases and development of
these plants was followed from May until late August.
Unless mortality took place within the first three weeks
of the season, plants which died were not replaced.
Individual leaves on each culm were periodically
evaluated for physical condition (green and developing,
damaged tip, brown and dead etc.) (Appendix A). As long as
a culm had any leaves which contained some green tissue,
31
the culm was classified as being alive. As these culms
were all roughly the same size and possessed roughly the
same number of leaves early in the season, they are
assumed to be approximately the same age (from date of
emergence in early spring). T h u s , all culm mortality
calculations are based on observations of an early season
cohort of culms. Results are expressed as the percentage
of this cohort which remain alive at a given date.
On 18 July, 1978 a severe hailstorm knocked down many
of the culms at the Agropyron smith!! site. Some culms
never recovered from the shock of the mechanical damage,
others were lost from the sample se t , and presumed dead.
This culm death fell nonuniformly on the treatments (six
in the wet plot, four in the 12 mm, zero in the 6 mm, and
two in the control) and could have no correlation to
supplemental water. For this reason, these deaths have
been removed from the analysis.
Statistical interpretation of the results of this
survey were made using a Chi-square analysis,
testing the
expectation of equal mortality occurring within each
irrigated treatment and the non-irrigated plots.
Expectations were weighted to compensate for the different
sample sizes involved.
32
RESULTS
Agropyron smithii Control
In 1977 , mortality of Agropyron culms did not begin
until mid July (Table 4). Leaf water potential had fallen
below -1.4 MPa 21 days earlier (Figure I). After the onset
of mortality, culm death continued steadily for the rest
of the season. By the end of August, 50% of the original
20 culms had died.
The summer of 1978 was wet (Table 2). Leaf water
potentials in the control plots first fell below -1.0 MPa
only in early August. This may have affected the control
plot mortality because while slight culm mortality did
occur in July and August, end-of-season survival was only
79% of the original early season cohort.
In 1979, as in 1977, mortality began in mid July
(Table 5); about three weeks after leaf water potentials
fell below -1.4 MPa. Culm death continued at a high rate
until early August when the culm deaths slowed. Although
leaf water potentials continued to be below -2.0 MPa, culm
death did not continue. This mortality may have been
prevented by a large rain event (2.8 cm) which took place
in the week of August 5-12. While the water was
33
insufficient to reduce leaf water stress (as measured), it
may have inhibited culm death.
Table 4. Agropyron smithii culm survivorship as % of
original culms alive at a given date for four
irrigation treatments, 1977 - 1978. Single (*)
and double (**) asterisks indicate, respectively,
p< .05 and p< .01 probability of significant
differences existing between treatments and the
control plots.
Date
Control
6_ mm
1977
5-24-77
6-01-77
6-07-77
6-15-77
6-24-77
7-05-77
7-12-77
7-19-77
7-26-77
8-02-77
8-10-77
8-16-77
8-25-77
8-30-77
100
100
100
100
100
100
100
90
85
85
75
75
50
50
100
100
100
100
95
95
95
95
95
95
95
95
95**
90**
1978
5-24-78
6-06-78
6-15-78
6-20-78
6-27-78
7-05-78
7-11-78
7-18-78
7-25-78
8-09-78
8-16-78
8-23-78
8-30-78
100
95
95
89
89
89
89
89
79
79
79
79
79
100
95
95
95
95
95
95
95
90
90
86
86
86
T reatment
12 mm
— ——
—— —
—
—
— ——
—
———
—
-—
———
—
—— —
___
—
100
100
94
94
94
94
94
94
88
81
81
81
81
Wet
100
100
100
100
100
100
100
100
100
100
100*
100*
100**
100**
100
100
100
100
100
100
100
100
93
86
86
86
79
34
Table 5. Agropyron smithii culm survivorship as % of
original culms alive at a given date for four
irrigation treatments (1979). Single (*) and
double (**) asterisks indicate, respectively,
p< .05 and p< .01 probability of significant
differences existing between treatments and the
control plots.
Date
Control
1979
5-14-79
5-21-79
6-05-79
6-12-79
6-20-79
6-26-79
7-03-79
7-11-79
7-18-79
8-02-79
8-09-79
8-14-79
8-24-79
100
100
100
100
100
100
100
80
65
30
30
30
20
6 nun
100
100
87
87
83
83
83
83
83
78**
78**
78**
70**
Treatment
12 mm
100
100
100
100
100
100
95
95
90
90**
90**
90**
85**
Wet
100
100
100
100
100
100
100
90
90
75**
70*
70*
60**
Agropyron smithii 6 mm
Even small amounts of water delayed culm death in dry
years. In 1977 and 1979, end-of-season mortality for the
plots receiving 6 mm of water was significantly less than
that of the control (90% vs. 50% in 1977 and 70% vs. 20%
in 1979) (Tables 4 and 5). In 1978, end-of-season culm
mortality did not differ significantly from the control.
Agropyron smithii 12 mm
During the two years when measurements were taken in
the plot receiving 12 mm water supplements, results
35
paralleled those observed on the 6 mm treatment.
a dry summer,
In 1979,
culm death was reduced relative to the
control (Table 5). End-of-season survival was 85% with
mortality starting in the first week of July, four weeks
after leaf water potential fell below -1.0 MPa (Figure I).
In wejt 1978, end-of-season culm death, was 81% in the 12 mm
treatment and did not differ significantly from the
control treatment .
Agropyron smlthii Wet
Addition of large amounts of water appeared to slow
the death of Agropyron culms relative to the control, but
did not have any greater effects than did applications of
lesser amounts of water (6 mm and 12. mm treatments).
During 1977, there were no culm deaths in the wet plot.
This did not differ significantly from the 6 mm. In 1978,
after compensation for hail mortality,
the end-of-season
culm mortality was indistinguishable from that of other
treatments. End-of-season culm deaths in the wet
treatment,
in 1979, were significantly lower than those in
the control, but were not higher than those in either the
6 mm or 12 mm treatments.
Bouteloua gracilis Control, 6 mm, and Wet
Bouteloua culm mortality was not influenced by
additional water (Table 6). End-of-season culm mortality
36
was slightly higher in all treatments during (wet) 1978
than in (dry) 1979. In both years, wet treatment end-ofseason culm mortality was 10% higher than in the control.
Table 6. Bouteloua gracilis culm mortality expressed as %
of original culms alive at a given date for three
irrigation treatments. No significant differences
were found between treatments and the control
plots.
Date
Control
Treatment
6 mm
Wet
1978
6-05-78
6-14-78
6-21-78
7-03-78
7-12-78
7-26-78
8-02-78
8-08-78
8-15-78
8-29-78
100
100
100
100
100
90
83
80
80
60
100
100
100
100
100
80
77
77
73
63
100
100
100
100
93
70
67
60
60
50
1979
6-13-79
6-19-79
6-27-79
7-05-79
7-11-79
7-17-79
8-08-79
8-21-79
100
100
100
100
100
100
97
90
100
100
100
100
100
100
97
97
100
97
97
97
97
97
87
80
37
DISCUSSION
The term mortality in this section does not have the
same meaning it would have in a discussion of animal
mortality. By culm death I refer to the presumptive death
of above-ground tissue; but this is only a fraction of the
plant. Roots are a major component of plant biomass and
this tissue may or may not die depending on the severity
and duration of the adverse conditions. What has been
called death (of above-ground tissue) may also be seen as
a retreat from harsh conditions on the surface.
Through the years of the experiment, conditions did
not remain constant. Culm density varied significantly in
many instances (Section I) and increased culm density has ,
been shown to be linked with increased culm mortality
(Kays and Harper 1974, Antonovics & Levin 1980 pg. 418).
Mortality in Agropyron smithii was slowed by small
amounts of supplemental water and slowed equally by large
amounts. This effect was most noticeable in dry summers
(1977,
1979).
The decreased culm mortality of Agropyron smithii in
wet treatments is expected,
since physiological water
stress was essentially eliminated for these plots. What
was not expected was that culm mortality would be greatly
.38
reduced by treatments (6 and 12 mm) which fell short of
eliminating water stress as conventionally measured.
In the case of Bouteloua gracilis, culm mortality was
unaffected by supplemental water. In the control plot
climatic conditions which provided a more mesic
environment (1978) resulted in an end-of-season mortality
which was higher than the dry year (1979).
39
CONCLUSIONS
Agropyron smithii
In dry summers (1977 and 1979), small amounts of
supplemental water were sufficient to reduce culm
mortality. During wet years (1978) mortality did not
differ significantly among control, light (6 and 12 mm),
and heavy treatments (wet).
Bouteloua gracilis
Bouteloua culm mortality did not appear to be
influenced by either large or small amounts of
supplemental water. During (wet) 1978, culm mortality was
slightly higher than in (dry) 1979 for all treatments. If
anything, the data suggest that supplemental water may
actually increase the death rate of Bouteloua gracilis
culms.
40
PART III
Effect of Supplemental Water on Leaf and Culm
Morphology In Agropyron smith!! and
Bouteloua gracilis
41
INTRODUCTION
The ability to capitalize on transitory periods of
favorable conditions is an important component of an
organism's success in a variable environment. For plants,
given their limited behavioral repertoire, this is
expressed in modification of physiology and morphology.
These changes must be flexible in magnitude to accommodate
a changing environment. Plastic morphological changes may
include alteration in I.) numbers of shoots, leaves, and
flowers, 2.) the size of vegetative parts, and 3.) degree
of hairiness (Bradshaw 1965).
This section describes plastic responses of two
economically important grasses (Agropyron smlthii and
Bouteloua gracilis) to water supplements such as might
result from minor climatic changes,
cloud seeding, light
irrigation, or heavy regular irrigation. The morphological
characters include culm height, number of leaves per culm,
number of leaf nodes on each culm, maximum length of
leaves,
and total length of green leaf tissue per culm.
Water stress generally causes reduction of growth.
Stunted plants have smaller stature, and fewer and smaller
leaves (Kramer 1969). Since supplemental water can
alleviate drought stress,
the results expected from adding
water are that by the season's end the plants would be
42
taller, have more nodes, have more and larger leaves on
each culm, and carry more green tissue than control
plants.
43
METHODS
In order to study water-induced morphological
modifications of Agropyron smithii and Bouteloua gracilis,
field plots were treated with four levels of irrigation.
The realized water stress for plants within the plots was
monitored (Figures I & 2) to document the effectiveness of
the treatments; and plant morphology within the study
plots was measured periodically with nondestructive
methods.'
Early in the spring, culms selected as representative
of the plot were chosen at regular intervals along a 2 x
14 m strip and tagged with cotton thread. Through the
summer, individual leaves on each culm were periodically
evaluated for physical condition (green and developing,
damaged tip, brown and dead) (Appendix A). As these culms'
were all roughly the same size and possessed roughly the
same number of leaves early in the season, they are
assumed to be approximately the same age (from date of
emergence in early spring). All culm morphology
calculations are based on this early season cohort of
culms.
As long as a culm had any leaves which contained some
green tissue the culm was classified as being alive. Leaf
lengths were measured as the distance from the
44
outstretched apex of the leaf , along the blade and sheath,
to the ligule of the next lower leaf. Culm height wasmeasured as the distance from the ground to the tip of the
uppermost outstretched leaf. For a more complete
description of leaf evaluation methods and data
management, see Appendices A and B .
In the data analysis the following conventions were'"'
used: I.) Mean culm height was calculated by summing the
heights of all plants alive at the sampling date and
dividing this value by the number of plants. 2.) Mean
number of leaves/culm was calculated by counting the
number of green leaves on each culm, summing these counts
and dividing the sum by the number of plants alive at the
sampling date. 3.) Maximum leaf length was calculated by
finding the largest length recorded that season. Only
those maxima which occurred at least one observation
before season's end were included. Similarly,
only those
maxima which occurred at least one observation before the
death of the plant were used for analysis. This allowed
confidence that these values were a measurement of the
maximum potential leaf length and not biased by culm
mortality or seasonal effects. Leaf conditions other than
death (brown and lost leaf tips for example) were not
considered as leaf mortality. 4.) Mean number of nodes per
culm was' calculated by counting the number of nodes on
each live culm and dividing by the number of culms alive
45
at the sampling date. 5.) Mean total length of green
tissue per culm was calculated by summing the lengths of
all leaves with green tissue on a given culm. These sums
were totaled and the grand sum divided by the number of
plants alive.
Culm height measurements were made on Agropyron
and
Bouteloua plants in 1978 and 1979. Leaf condition
measurements were made on Agropyron and Bouteloua plants
for the years 1977,
1978, and 1979. Leaf length
measurements were made on only Agropyron plants for the
same three years.
The results of these calculations were evaluated by
comparing results from each irrigation treatment with the
control results, using Student's t comparison of weighted
means from unequal sample sizes and unequal sample
variances (Snedecor and Cochran 1980 chapter 12).
46
RESULTS AND DISCUSSION
This section is divided into five parts, each
concerned with a single morphological character. These are
culm height, mean number of nodes per culm, mean number of
green leaves per culm, mean length of green leaf tissue
per culm, and maximum leaf length.
Culm Height
By season's end, large amounts of supplemental water
(wet treatments) made culms taller in both the wet year
(1978) and the dry year (1979). Smaller water supplements
had only slight effects.
1978. In a wet year (1978) (Table 2), culm height for
Agropyron smithii in the wet plots remained
indistinguishable from the control until early June (Table
7). After that, time, culm height increased rapidly in the
wet plot, with maximum heights being attained in mid-July.
At maximum height, Agropyron culms in the wet plots were
55 cm (or 47%) taller than the control culms. Any decrease
in mean culm height shown over the latter half of the
field season is due to mortality of taller culms during
this period.
47
Table 7. Mean culm height for Agropyron smithii in
1978 & 1979. Comparisons are made against the
control plot only. Single asterisk (*)
indicates p< .05, double asterisks (**)
indicate p< .01 .
Date
1978
5-02-78
5-10-78
5-16-78
5-24-78
6-06-78
6-15-78
6-20-78
6-27-78
7-05-78
7-11-78
7-18-78
7-25-78
8-09-78
8-10-78
8-16-78
8-23-78
8-30-78
Control
127.3
138.0
151.7
184.3
235.5
281.0
308.7
341.5
370.2
381.4
384.9
364.6
—-—
364.6
364.6
Treatment
6 mm
12 mm
109.2
127.6
141.2
180.8
236.6
270.6
308.0
336.7
343.8
351.8
353.3
358.2
358.2
117.2
131.3
139.1
189.4
257.1
324.2
357.3
395.8
427.1
440.5
443.9
400.5
407.0
407.0
407.0
407.0
Wet
133.5
152.2
172.6
216.4*
294.6**
377.I**
417.5**
478.7**
533.4**
551.I**
564.7**
552.2
546.8**
546.8
546.8**
565.8**
1979
5-14-79
5-21-79
6-05-79
6-12-79
6-20-79
6-26-79
7-03-79
7-11-79
7-18-79
8-02-79
8-09-79
8-14-79
8-24-79
116.7
138.5
173.6
185.1
189.5
190.2
190.2
187.6
192.3
201.0
204.7
204.7
211.2
105.5
131.6
159.9
168.2
174.1
174.3
174.3
174.3
174.3
183.8
178.2
178.2
182.8
122.8
142.6
170.9
175.3
176.3
176.9
179.8
179.6
185.4
185.4
185.6
185.6
186.6
139.7**
163.0**
230.4**
269.0**
316.6**
354.8**
379.4**
406.5**
407.8**
385.7**
384.0**
382.0**
390.7**
48
Table 8. Mean culm height for Bouteloua gracilis in
1978 & 1979. Comparisons are made against the
control plot only. Single asterisk (*)
indicates p<.05, double asterisks (**)
indicate p < .O l .
Date
1978
Control
6-05-78
6-14-78
6-21-78
7-03-78
7-12-78
7-26-78
8-02-78
8-08-78
8-15-78
8-29-78
89.6
96.7
101.4
140.6
156.5
150.4
147.9
142.0
141.9
106.4
Treatment
6 mm
95.2
103.8
109.8
136.7
163.8
165.9
167.3
167.3
171 .5
174.2
Wet
129.5**
145.6**
148.6**
169.2
184.6
154.0
156.9
144.1
144.5
152.9
1979
6-13-79
6-19-79
6-27-79
7-05-79
7-11-79
7-17-79
8-08-79
8-21-79
96.1
101.0
101.3
102.8
102.1
103.0
102.2
100.6
104.3
110.2
111.4
115.2
116.3
117.1
122.8*
122.9*
140.8**
149.3**
155.9**
166.3**
178.5**
190.7**
202.4**
207.5**
Mean height of culms in the 12 mm treatment showed
intraseasonal dynamics similar to those in the wet plots.
At no time did the mean culm height differ significantly
from the control plot heights (Table 7).
Water treatments of 6 mm also caused no significant
increases in culm height relative to the control in 1978.
Culm growth proceeded until early July, after which growth
slowed and then stopped. Mean culm height was not
significantly less than the control after mid-June.
49
In 1978, Bouteloua gracilis was not significantly
affected by heavy irrigation (wet treatments). This may be
due to the wet year and the xerophytic nature of
Bouteloua. Through June plants growing under the wet
treatment were significantly taller than the controlplants (Table 8), but the differences disappeared by early
July.
1979. In (dry) 1979, the wet treatment produced
larger effects upon culm height relative to the control in
Agropyron smithii (Table 7) than it did in 1978. However,
the maximum culm height was not as, great as that in 1978.
This was probably due to the lower humidity of 1979.
Interseasonal dynamics were very similar to 1978, with
maximum height attained in mid-July. At that time, wet
plot culms were 112% taller than the control plot culms.
Neither 12 mm nor 6 mm plot culm heights were
significantly different from control in 1979. At season's
end, 6 mm and 12 mm plot culms were roughly 15% shorter
than those of the control. This figure was not
statistically significant and no explanation is offered
for the observation.
In 1979, Bouteloua gracilis height responded to heavy
watering (Table 8). By the end of August, wet treatment
culms were 107% taller than those in the control plots.
Small amounts of water (6 mm) in this dry summer (1979)
caused no significant effects on culm height.
50
Culm Height Conclusions. Culm heights in Agropyron
and Bouteloua can respond to supplemental water. Large
amounts of water made the culms taller . This effect was
more pronounced in dry years than in wet years. Small
amounts of supplemental water had little effect on culm
height for either species in both moist and dry years.
Mean Number of Nodes per Culm
Nodes in grass plants are produced at the junction of
each leaf with the stem. While individual leaves may die
or detach, the nodal structure will remain as long as the
stem is intact. Since Agropyron smithii and Bouteloua
gracilis grow from apical meristems,
the number of nodes
on a culm is a record of the cumulative number of leaves
produced by the culm since emergence in early spring.
Agropyron smithii. In the dry summers of 1977 and
1979, by the end of the year, mean numbers of nodes on
each culm of Agropyron smithii were highly influenced by
supplemental water (Table 9). In wet 1978, however, the
end of season mean node number differed little among
treatments.
In 1977, start-of-season node number in the 6mm and
wet plots was not statistically different from the control
(Table 9). By the end of the year, mean node number was
6.7 in the control, 7.6 in the 6 mm, and, 11.2 in the wet
plot.
51
Due to the early season rainfall in 1978 (Table 2 and
Figure I), the mean node number did not differ
significantly from the control for any irrigation level
except the wet treatment during 1978 . End-of-season means
were 9.1 ±.27 S.E. for the control and 11.5 ±.61 S .E . for
the wet plot (Table 9).
In 1979, node number at the beginning of the season
did not differ among the treatments (Table 9). With the
onset of drought stress in early June, the rates.of leaf
production were slowed in all but the wet plot. The mean
number of nodes on each culm differed between the wet
plots and all- other treatments after that time. End-ofseason values for the four treatments were 6.5 ±.29 S.E.
in the control, 6.8 ±.19 S.E. in the 6 mm plot, 6.7 ±.33
S.E. in the 12 mm plot, and 8.3 ± 5.4 S.E. in the wet
(Table 9)-.
52
Table 9
Date
1977
5-24-77
6-01-77
6-07-77
6-15-77
6-24-77
6-28-77
7-05-77
7-12-77
7-19-77
7-26-77
8-02-77
8-10-77
8-16-77
8-25-77
8-30-77
Mean nodes per culm in Agropyron smithii during
1977 - 1979. Comparisons are made against the
control plot only. Single asterisk (*)
indicates p < .05, double asterisks (**)
indicate p < .01.
Control
Treatment
6 mm
12 mm
4.3
5.0
5.4
5.4
5.4
—
5.4
5.7
5.8
5.8
6.1
6.0
6.1
6.4
6.7
4.6
4.8
5.3
5.3
5.3
———
5.4
6.1
6.3
6.4
6.5
6.9*
7.2**
7.4*
7.6
3.7
4.1
4.3
5.1
6.0
6.7
7.1
7.7
8.2
8.4
8.6
-—
9.0
—— —
9.1
9.1
3.5
3.7
4.1
4.9
5.9
6.6
—
7.4
8.2
8.6
— ——
9.1
9.3
9.4
9.5
— ——
——— —
— —— —
—
— —— —
—
— — ———
— —— —
— —— —
— —— —
— —— —
——— —
— — ———
— —— —
— — ——
—
Wet
4.6
5.1
5.9*
6.8**
7 .I**
7.9
8.3**
9.0**
9.4**
9.9**
10.1**
10.7**
10.8**
11.0**
11.2**
1978
5-02-78
5-10-78
5-16-78
5-24-78
6-06-78
6-15-78
6-20-78
6-27-78
7-05-78
7-11-78
7-18-78
7-25-78
8-09-78
8-16-78
8-23-78
8-30-78
3.5
3.9
4.1
4.9
5.7
6.4
6.8
7.2
7.8
8.2
8.5
9.2
9.5
9.5
9.5
9.7
3.4
3.9
4.4
5.0
5.9
6.5
6.7
7.3
7.8
8.3
8.6
9.4
10.3*
10.7
11.0*
11.5**
53
Table 9 Continued
Date
1979
_______________ Treatment
Control
6 mm
12 mm
5-14-79
5-21-79
6-05-79
6-12-79
6-20-79
6-26-79
7-03-79
7-11-79
7-18-79
8-02-79
8-09-79
8-14-79
8-24-79
3.7
4.3
5.3
5.7
5.8
5.8
5.8
5.9
6.0
6.2
6.2
6.2
6.5
3.7
4.2
5.1
5.6
6.0
6.0
6.0
6.0
6.0
6.4
6.6
6.7
6.8
3.7
4.1
5.0
5 .I**
5.2*
5.5
5.6
5.8
5.9
6.4
6.6
6.7
6.7
Wet
3.5
4.1
5.2
5.7
6.3*
6.5**
6.9**
7.2**
7.4**
7.7**
7.8*
8.0**
8.3*
Bouteloua gracilis. Mean node number for Bouteloua
Rracilis was little affected by supplemental water in
either the dry or the wet year (Table 10).
In 1978 initial node numbers were 4.4
13 S .E . , 4.7
j+. 10 S .E . , and 4.2 _+. 12 S .E . for the control , 6 mm, and
wet plots respectively. End-of-season values were 6.6 ± . 28
S .E . , 6.7 ;+. 34 S.E., and 7.2 _+. 35 S.E. for the same three
treatments.
During the dry year (1979), mean node numbers for
Bouteloua plants differed little from the wet year of 1978
(Table 10). Initial values did not differ among the
treatments; end-of-season values were only slightly higher
for the wet plots (Table 10).
54
Table 10. Mean nodes per culm in Bouteloua gracilis during
1978 - 1979. Comparisons are made against the
control plot only. Single asterisk (*)
indicates p < .05, double asterisks (**)
indicate p < .01.
Date
1978
Control
6-05-78
6-14-78
6-21-78
7-03-78
7-12-78
7-26-78
8-02-78
8-08-78
8-15-78
8-29-78
Treatment
6 mm
Wet
4.4
5.0
5.4
5.7
5.9
6.1
6.1
6 .I
6.2
6.6
4.7
5.3
5.7
6.1
6.1
6.4
6.5
6.5
6.6
6.7
4.2
4.9
5.2
5.5
5.8
6.0
6.3
6.4
6.6
7.2
5.0
5.3
5.7
5.8
5.8
5.9
6.1
6.2
4.8
5.0
5.5
5.9
6.0
6.0
6.2
6.2
4.6*
5.0
5.5
5.9
6.1
6.3
6.6
7.0*
1979
6-13-79
6-19-79
6-27-79
7-05-79
7-11-79
7-17-79
8-08-79
8-21-79
Mean Number of Leaves per Culm
The number of leaves sustained by an individual culm
might be expected to increase with water supplements which
relieve water stress. The net number of leaves borne on a
culm may be constant through a season, while, at the same
time, considerable turnover in leaves may occur through
leaf emergence and death. Constraints on numbers of leaves
sustained by a culm may be imposed by desiccation. Numbers
55
of green leaves may have to be limited in times of water
stress to prevent desiccation of the entire plant. If
water stress were eliminated, would the plant continue to
add leaves through the year?
Initial Values. Initial numbers of leaves on culms
did not differ significantly among the treatment plots for
either Agropyron smithii or Bouteloua gracilis. Mean
numbers of Agropyron leaves/culm were 3.5, 3.9, and 3.7
respectively for the control,
May,
6 mm and wet plots on 24
1977 (Table 11). The pre-treatment mean leaf numbers
for Bouteloua gracilis are 3.9, 4.2, and 4.1 (Table 12).
Mean leaf numbers for the season's first observation were
not different among the various treatments in any of the
years studied for either species. Since perennating tissue
of these plants dies back each winter, pre-irrigation
growth conditions are similar. There apparently is no
cumulative effect of supplemental water on accelerating
early, growth in leaf numbers.
56
Table 11. Green leaves/culm in Agropyron smithii during
1977, 1978, and 1979. Comparisons are made
against the control plot only. Single asterisk
(*) indicates p < .05 , double asterisk (**)
indicates p < .01.
1977
Date
5-24-77
6-01-77
6-07-77
6-15-77
6-24-77
6-28-77
7-05-77
7-12-77
7-19-77
7-26-77
8-02-77
8-10-77
8-16-77
8-25-77
8-30-77
Control
Treatment
6 mm
12 mm
Wet
2.8
3.0
2.7
2.5
2.6
2.7
2.8
3.2
3.5
3.9
4.2
4.3
3.8
3.5
—— —
3.1
3.7**
3.9**
3.8**
3.8**
4.2**
4.4**
4.3*
4.6*
— —— —
— ——
— — ——
____
—— ——
— —— —
— —— —
— —— —
— —— —
————
— —— —
————
— —— —
—— — —
— ---
3.8
3.8
4.5
5.2**
5.3**
6.1
6.3**
6.8**
7.0**
7.4**
7.4**
7.9**
7.9**
8.0**
8.2**
3.3
3.0
3.2
3.6
4.2
4.8
5.2
5.2
5.1
5.1
4.9
—— —
3.9
—
2.8
2.4
3.1
3.0
3.0
3.8
4.4
4.8
—
4.8
5.3
5.4
—
4.8
3.7
3.4
3.0
-—
3.3
3.1
2.9
3.8
4.2
4.7
5.0
4.9
5.2
5 .I
5.2
5.0
4.0
3.4
3.2
2.6
3.2
3.1
3.1
3.4
3.8
4.2*
4.2**
4.4*
4.4
4.6
4.7
5.2
5.3*
5.5
5.8**
5.4**
3.5
4.1
4.1
3.6
3.5
——
1978
5-02-78
5-10-78
5-16-78
5-24-78
6-06-78
6-15-78
6-20-78
6-27-78
7-05-78
7-11-78
7-18-78
7-25-78
8-09-78
8-16-78
8-23-78
8-30-78
57
Table 11. Continued
Date
1979
____________
Treatment
12 mm
Control
6 mm
5-14-79
5-21-79
6-05-79
6-12-79
6-20-79
6-26-79
7-03-79
7-11-79
7-18-79
8-02-79
8-09-79
8-14-79
8-24-79
3.6
3.8
4.3
4.6
4.0
3.6
2.8
2.3
2.2
2.0
2.0
I .7
1.8
3.7
3.9
4.2
4.2
4.0
3.4
3.3
2.9*
2.6
3.0**
3.1**
3.1**
3.1
3.6
3.6
3.7*
3.6**
3.0**
3.0*
2.8
3.1**
3.1**
3.3**
3.4**
3.5**
3.5*
Wet
3.5
3.6
3.8*
3.8**
4.1
3.9
4.0**
4.2**
4.3**
4.3**
4.2**
4.4**
4.4**
1977 Agropyron smithii. In 1977, numbers of leaves
per culm stayed within fairly narrow limits for the
control and 6 mm treatments, ranging from 2.5 to 4.1
leaves for the control and 3.1 to 4.6 for the 6 mm
(Tablell). These numbers are far lower than those in the
wet plot. Wet plot mean culm numbers increased throughout
the 1977 season. End-of-season mean number of leaves was
8.2 for plants in the wet plot, 4.6 for plants in the 6
mm, and 3.5 for plants in the control. Comparison of means
for these end-of-season figures show leaf numbers for both
the 6 mm and the wet treatments to be significantly larger
than those in the control.
1978 Agropyron smithii. Early in 1978, mean leaf
numbers were very similar for all four treatments (Table
58
11). Leaf numbers increased for all treatments until the
middle of the season. After July 10, mean leaf numbers
decreased for the control, 6 mm and 12 mm plots. The
number of leaves per culm in the wet plot continued to
increase through the year, reaching a maximum number of
5.7 leaves on August 23. This number was significantly
greater than that of the control. End-of-season mean leaf
numbers did not differ among the control, the 6 mm, and 12
mm plants.
1979 Agropyron smithii. In 1979, initial leaf numbers
were similar for all four treatments (Table 11). Wet plot
leaf numbers increased only slightly during the course of
the year, with end-of-season mean numbers of 4.4
leaves/culm. The 6 mm and 12 mm treatments had end-ofseason mean leaf numbers of 3.1 and 3.5 leaves/culm.
Initial leaf numbers in the control increased during the
first 3 weeks of May and after that decreased for the rest
of the season. End-of-season leaf numbers for the control
were 1.7 leaves/culm. All three irrigation treatments had
numbers significantly higher than the control during the
last half of the season (beginning in the last week of
July).
59
1978 Bouteloua gracilis. Mean number of leaves/culm
did not differ significantly among the three treatments in
1978 (Table 12). In general, the number of leaves
decreased over the season.
Table 12. Green leaves/culm in Bouteloua gracilis during
1977 (start-of-season only), 1978, and 1979.
Comparisons are made against the control plot
only . Single asterisk (*) indicates p < .05,
double asterisks (**) indicate p < .O l .
Date
1978
Control
5-26-77
6-05-78
6-14-78
6-21-78
7-03-78
7-12-78
7-26-78
8-02-78
8-08-78
8-15-78
8-29-78
Treatment
6 mm
Wet
3.9
4.3
4.3
4.4
4.4
4.5
3.9
3.6
3.4
3.1
3.2
4.2
4.6
4.7
4.8
4.8
4.4
3.7
3.7
3.5
3.1
3.1
4.1
4.2
4.1
3.9*
3.8
3.9
3.5
3.6
3.4
3.2
3.4
4.6
4.4
4.3
4.1
3.9
3.5
3.3
2.9
4.6
4.6
4.7
4.7**
4.3*
4.0**
3.5
3.0
4.3
4.5
4.5
4.7**
4.6**
4.3**
4.2**
4.0**
1979
6-13-79
6-19-79
6-27-79
7-05-79
7-11-79
7-17-79
8-08-79
8-21-79
1979 Bouteloua gracilis. This year was much drier
than the previous year. Mean leaf numbers per culm
decreased through the season in the control and 6 mm
treatments, while the wet treatment numbers changed little
60
from start-of-season values (Table 12). End-of-season mean
leaves/culm in the wet treatment were significantly higher
than the control treatment with 4.0 vs. 2.8 leaves/culm
(Table 12).
Mean Leaf Number Conclusions
Supplemental water increased the mean number of
leaves per culm of Agropyron smithii in dry years. This
was true for small amounts of water as well as large ones.
In wet years, light addition of supplemental water did not
increase the leaf numbers, but heavy irrigation did.'
For Bouteloua gracilis, supplemental water increased
the mean leaves/culm only if the year was dry and large
amounts of irrigation were applied.
In no case was the early season leaf number
morphology changed by addition of water. All plants showed
the same numbers of leaves early in the spring regardless
of previous water treatment. This is. in contrast to
Agropyron smithii culm density which showed strong
interseasonal cumulative effects (Section I).
Maximum Leaf Length- Agropyron smithii
As leaf surface area is both critical to net
photosynthetic gain and desiccation potential, it was
anticipated that supplemental water would make the leaves
of Agropyron smithii larger. To eliminate the effects of
61
differential rates of leaf maturation on size
measurements, only leaves at maximum elongation were
considered. As leaves are also 1expected (Robson 1973) to
differ in maximum size based on their ranking in acropetal
order (from the base of the plant upward), it was
necessary to compare leaves which were of equivalent rank
order.
Maximum lengths of leaves were calculated by scanning
the records for maximum leaf lengths for leaf positions
#3, #4, and #5 (Table 13). Leaves in these ranks generally
were still growing at the start of the field season but
had enough time before summer's end to reach maturity. In
cases where insufficient numbers of undamaged leaves
existed to make valid estimations for a particular rank
order of leaves, the rank order was dropped from the
survey.
Because leaves mature in acropetal order, in certain
cases the leaf in position #3 was fully developed before
treatment effects could be felt, and effects of the
supplemental water are not seen. For example this was
probably the case for the wet plot in 1977 (Table 13).
In 1977, the only leaf position which differed from
the control plot's leaves of the same rank was position #5
in the wet plot. Its mean length was 26% longer than that
found in the control (Table 13).
62
In 1978, leaves in the 4th and 5th positions in the
wet plot were significantly longer (131% and 121%
respectively) than those in the control (Table 13). The
mean length of the 4th leaf in the 12 mm plot was also
significantly larger . Leaves in the 6 mm plot were not
appreciably larger than control leaves in any year.
Table 13. Mean maximum length (mm) of Agropyron smithii
leaves subjected to four irrigation treatments.
Comparisons are made against the control plot
only. Single asterisk (*) indicates p< .05,
double asterisks (**) indicate p < .Ol .
Treatment
12 mm
6 mm
Wet
Year
Control
#3
#3
#3
1977
1978
1979
143
Ill
109
101
#4
#4
#4
1977
1978
1979
150
135
162
123
178*
123
196*
170*
#5
#5
#5
1977
1978
1979
119
185
121
120
171
122
200
121
150*
223*
195*
Leaf #
—
138
-
-
138
-
-
144*
—
—
In 1979, only the wet plot had leaves which were
larger than the control leaves. In this year,
leaves
all three positions grew larger. The percentage of
additional growth was of the same general order as those
in previous years at 130%,
control leaves.
127%, and 163% longer than the
63
Heavy irrigation clearly promotes larger leaves on
Agropyron plants. This effect was observed in both wet and
dry years. Lighter water supplements had less dramatic or
negligible effects although there is some indication
(1978, 12 mm treatment) that smaller amounts of water
might marginally increase the size of leaves.
Mean Green Tissue per Culm
The total length of green tissue on a culm is
determined by both the number of leaves on each culm and
the size of those leaves. Since it has been shown that the
number of leaves per culm increased as a result of
increased water and that selected leaves had larger leaves
when large amounts of supplemental water were added to a
natural grassland- it is reasonable to conclude that the
total length of green tissue per culm should also have
increased as a result of adding large amounts of water.
1977.
Wet plots showed dramatic changes in the
morphology of Agropyron smithii culms this year (Table
14). Total length of green tissue, which is highly
correlated with total photosynthetic area (Newbauer 1985),
increased over the entire year (Table 14). By the end of
the season, wet treatment culms had a green tissue length
of 1135 mm/culm. This was 308% of the control plants at
the same date. Large increases in total culm leaf length
were found in the first two observations of the 6 mm and
64
Table 14. Green leaf tissue length (mm/culm) of
Agropyron smithii 1977-1979. Comparisons are
made a gains!: the control plot only. Single
asterisk (*) indicates p < .05, double asterisk
(**) indicates p < .01.
Date
1977
5-24-77
6-01-77
6-07-77
6-15-77
6-24-77
6-28-77
7-05-77
7-12-77
7-19-77
7-26-77
8-02-77
8-10-77
8-16-77
8-25-77
8-30-77
_____________ Treatment_____________
Control
6 mm
12 mm
Wet
298
394
467
417
415
312
378
460
421
388
—
—
—
—
—
—
—
—
—
—
—
—
—
334
357
300
278
279
284
302
350
368
—
349
397
420**
401**
409**
430**
455**
443
471*
236
272
307
397
541
681
774
805
851
881
850
198
304
191
419
548
681
-----
—
—
—
—
—
—
—
—
-----
—
—
—
—
—
—
—
—
—
335**
435*
538*
656**
734**
830
910**
959**
1004**
1052**
1068**
1097**
1110**
1123**
1135**
1978
5-02-78
5-10-78
5-16-78
5-24-78
6-06-78
6-15-78
6-20-78
6-27-78
7-05-78
7-11-78
7-18-78
7-25-78
8-09-78
8-16-78
8-23-78
8-30-78
-----—
619
—
410
328
—
—
—
757
856
859
—
785
557
485
413
—
—
—
221
256
385
537
712
746
812
893
896
940
890
668
532
500
372
284
298
323
429
588
735
793
843
905
933
980
1040
1002**
917
982**
867**
65
Table 14 Continued
Date
1979
________________ Treatment
Control
6 m
12 mm
5-21-79
6-05-79
6-12-79
6-20-79
6-26-79
7-03-79
7-11-79
7-18-79
8-02-79
8-09-79
8-14-79
8-24-79
281
392
437
422
390
310
246
238
235
241
192
205
242
364
386
384
341
327
279
255
301
308
306
301
282
361
378
326*
315
301
316
333*
342
345
354*
361
Wet
328
484**
530*
607**
633**
658**
730**
748**
716**
693**
709**
726**
control plots. Plants in the latter two treatments had
leaves which first experienced water stress greater than
-0.8 MPa on 7 June and after that the mean green tissue
length decreased for both these treatments. By the end of
the season, control plot green leaf length was little
changed from the beginning of the season and had exhibited
a narrow range of values. This was also true for the 6 mm
plots. Leaf length on 6 mm treatment culms, however, was
greater than that on control plants by the end of the
season.
1978.
All four treatments showed marked increases in
green tissue length for the first half of 1978- the period
of abundant natural water. Water stress first exceeded
-0.8 MPa on 24 July for the 12 mm, 6 mm, and control
treatments. After the end of June, the wet plot culms
66
continued -to increase their green tissue length for
several weeks while the accumulation of green tissue under
other treatments slowed. After 19 July and for the last
half of the summer, mean culm leaf length decreased. The
response of these plants to ample water resources was a
rapid increase in photosynthetic area and the response to
water stress was a decrease in green tissue.
In 1978 , there was no significant response over the
control by the 6 mm and 12 mm treatments. Wet plots showed
significant responses after I July when water stress
appeared in the other plots.
1979.
In the third year of the treatment, early
season increases in green tissue were identical to those
of previous years. Onset of water stress was much earlier
in this year, however, and for all treatments but the wet,
maximum leaf length occurred on 11 June. Wet plots
continued to increase green tissue until mid.July. Plants
in the 6 mm, 12 mm, and control plots lost mean green leaf
tissue/culm after 11 June. The rate of decrease slowed in
the 6 mm and 12 mm plots by the third week in Ju l y , when
culms in all three irrigated treatments had significantly
more green tissue than the control.
Green Tissue Length Conclusions. Supplemental water
did increase the photosynthetic area of Agropyron smithii
culms. For those years in which summer rain was scarce,
67
large amounts of supplemental water produced culms with
much greater leaf area than culms which received'only
natural rainfall. Smaller amounts of supplemental water
had less dramatic results (Table 14).
Density Effects
Analysis of the effect of supplemental water on culm
morphology is complicated by the non-uniform density which
resulted from water supplements (Section I). Density has
been shown to affect plant morphology (Kays 1974).
Antonovics and Levin (1980 p .418) have shown that plants
will respond to medium densities with a reduction in
growth rate . In a study of this kind in which perturbation
of the plant community occurs at many levels, it quickly
becomes difficult to differentiate between effects which
are the direct result of relief from water stress and
those effects which are secondary to the treatment.
68
CONCLUSIONS
In general, the morphological changes which resulted
from supplemental water confirmed the expectations which
resulted from previous studies of water stress (Simpson
1981, Kramer 1969). Agropyron smithii grew taller with
addition of large amounts of water (wet treatments),
less
so with lighter irrigation treatments. Numbers and.size of
leaves on Agropyron culms was increased by large amounts
of water, smaller amounts had little effect. Total length
of green leaf tissue on culms of Agropyron smithii was
only affected by large amounts of irrigation.
Supplemental water had a much less marked effect on
Bouteloua gracilis. While culm height was affected by
irrigation, leaf numbers were not.
69
PART IV
Senescence, Emergence, and Maturation Rates in Leaves
of Agropyron smithii and Bouteloua gracilis
70
INTRODUCTION
Scope of the Study
A differential in the rates of leaf emergence and
senescence will result in increased or decreased numbers
of leaves found on a culm. Having shown that supplemental
water can produce morphological changes in the number of
leaves sustained on grass culms (Section III), records of
leaf condition and survival were analyzed to determine
relative magnitudes of the processes which resulted in the
change.
Investigation of emergence and senescence also has
the capacity to disclose turnover in a leaf population
which might otherwise go unnoticed. Simple counts of the
numbers of individuals present at a given time may answer
questions of density but will tell nothing of the dynamics
in the populations involved (Nobel e_t a_l. 1979). New
individuals may be emerging while older individuals are
disappearing and the net change in the community is zero.
Without knowledge of turnover rates a dynamic community
might appear static.
Both emergence and senescence have relationships to
plant physiology (Hsiao 1973, Webster 1973) and may
reasonably be expected■to be correlated with the water
71
status of plants. One might wish to correlate the rates of
senescence and emergence with the degree of water stress
experienced by the grass plant. One might also look for a
differential seasonal response in leaf emergence. Since
the energy invested in leaf production at the end of the
season might not have time to be repaid by the resulting
increased photosynthetic production , one might expect a
decrease in leaf emergence with season.
Other questions which present themselves at this
juncture are whether supplemental water would delay or
accelerate the developmental time of leaves.
These questions will be considered in this section.
Concepts Related to Leaf Dynamics
To view leaf emergence and senescence as independent
processes occurring simultaneously within a plant, it is
necessary to recognize the metameric nature of leaves.
Plant growth is largely indeterminate and the major
changes of plant form occur by varying the number of
modular units (leaves, buds, phytomers, tillers) which
make up a plant.
The practice of viewing a plant as an aggregate
population of leaves has a long history (White 1979).
Plant population biology as discussed by John Harper
(1977) has made use of the leaf population concept to
72
elucidate.processes (such as emergence and senescence
rates) in plant biology which required this approach.
A simple model of leaf population numbers has been
given by Harper (1977) which describes both the current
population size as well as the turnover rate of the
individual members. Here N q is the population size at time
0 and the subscripted variable N t+1- is the population size
after the next unit of time has elapsed.
.Nt + 1=NgfBirths-Deaths+Immigrants-Emigrants
In the case of leaves, with the few exceptions of
vegetative cloning,
either to immigrate or emigrate is to
die and the equation becomes an even simpler one of births
and deaths or emergence and senescence.
73
.METHODS
To determine how the dynamics of Agropyron smithii and
Bouteloua gracilis leaf populations react to supplemental
water, a field study was conducted in which natural rainfall
was augmented with irrigation. Leaves within the study plots
were measured periodically with nondestructive methods and
the degree of water stress was monitored (Figures I & 2).
General details of the experiment have been described in the
General Introduction under the sections "The Study Site" and
"Irrigation Regimen".
Rates of leaf emergence were calculated by counting the
number of leaves which had emerged since the last
observation and scaling the rates to the common unit of
number of leaves emerged per IOO culms per w e e k . Leaf
senescence rates were calculated in an analogous fashion.
Mean weekly rates of leaf emergence and senescence for a
particular year were calculated by averaging the scaled
weekly rates over a given field season.
To correlate the rates of leaf emergence and senescence
with plant water status it was necessary to pair the leaf
water potential observation with the interval over which the
emergence and senescence rates were affected. Because
reported leaf emergence and senescence data was the result
of processes which had happened since the previous
74
observation, leaf water potential data for a given date was
matched with the following emergence or senescence date. In
this way it was possible to test leaf water potential as a
predictor of leaf population dynamics.
Tests of seasonal trends in emergence and senescence
rates were made by segregating observations for the months
Ju n e , July, and August and comparing the mean rates measured
in these periods.
The time it took a newly emerged leaf to reach it's
maximum length was defined as the leaf's juvenile period. To
measure this interval:
I.) A date was selected which was
early enough in the year that leaves appearing then could
reach maturity. 2.) Data on all leaves which emerged at that
date were scanned, and the Julian date at which the leaf was
first recorded as being present was subtracted from the last
date it showed growth. This interval was called the juvenile
period of the leaf. Use of a particular cohort of leaves in
these calculations was necessary to remove the possible
confounding effects of season and weather on the duration of
the elongation process.
Leaf condition measurements, which form the basis of
the leaf emergence and senescence rates, were made for both
Agropyron smithii and Bouteloua gracilis. Leaf length
measurements, which form the basis for the leaf maturation
calculations, were made only for Agropyron plants.
75
RESULTS
Mean seasonal emergence and senescence rates will be
discussed first, comparing years, treatments and net
change brought about by these two opposing processes.
Second, two factors which may influence senescence and
emergence rates, seasonal effects and water potential, are
considered. Finally,
leaf's existence:
the duration of two phases of a
I.) the time from first emergence until
cessation of expansion, and 2.) the time from cessation of
growth until senescence, will be contrasted among
irrigation treatments.
Seasonal Averages of Emergence,
and Senescence Kates
Emergence and senescence rates were averaged over the
entire field season for each treatment. By doing this one
may show the mean rate of leaf production and leaf death
in each treatment for each year. Leaf turnover was
prorated to compensate for the uneven time elapsed between
observations. The rate variance was calculated as a
binomial distribution of plants which produced (or lost)
leaves since the previous observation.
Comparison of Treatments. Comparison (Student's t ) of
irrigated seasonal means with the control could not show
76
significant differences in emergence or senescence rates
in any instance for either Agropyron smithii or Bouteloua
gracilis (Table 15 and 16). This result was unexpected in
light of the significant differences between treatments
shown by mean leaf numbers in Section III.
Table 15. Mean weekly leaf emergence and senescence rates
for Agropyron smithii under four different
irrigation regimens in the summers of 1977-1979.
Rates are calculated as the mean number of
leaves emerged or senesced for 100 culms in a
one week interval. Difference between the two
rates is given.
Control
6 mm
12 mm
Wet
24 May-30 A u g . 1977
Emergence
Senescence
22.0
18.2
23.4
18.2
47.2
14.3
Difference
-3.8
+ 5.2
+ 32.9
Control
6 mm
12 mm
Wet
~~T~May-30 A ug. 1978
Senescence
Emergence
34.7
32.7
40.6
36.7
40.8
35.2
36.2
43.7
Difference
-2.0
-3.9
-5.6
+ 7.5
Control
6 mm
12 mm
Wet
14 May-24 A u g . 1979
Senescence
Emergence
40.6
18.2
27.9
16.5
22 .I
17.2
32.1
26.9
Difference
-22.4
-11.4
-4.9
+ 5.2
77
Table 16. Mean weekly leaf emergence and senescence rates
for Bouteloua gracilis under three different
irrigation regimens for the years 1978 to 1979.
Rates are calculated as the mean number of
leaves emerged or senesced for IOO culms in a
one week interval. Difference between the two
rates is given.
Control
6 mm
Wet
5 June-29 A u g . 1978
Emergence
Senescence
30.0
17.0
30.5
16.4
36.5
22.3
Difference
-13.0
-14 .I
-14,2
r
Control
6 mm
Wet
13 June-21 A u g . 1979
Emergence
Senescence
34.7
13.5
34.3
17.2
31.5
29.0
Difference
-21.2
-17.1
-2.5
Comparison of Years. Wet 1978 generally showed the
highest leaf emergence rates at both the Agropyron and the
Bouteloua field sites. The single exception was the wet
treatment rate in that year, which was slightly lower than
the 1977 rate. Dry 1979 exhibited the lowest emergence
rates. Given the relative natural moisture levels of these
two years, this is an indication of a positive correlation
between water and, emergence rates even though comparison
of irrigation treatment seasonal means could prove no such
correlation .
Senescence rates showed no comparable relationship
with natural rainfall. The year with the lowest rate was
1977 for Agropyron smithii, while wet 1978 and dry 1979
had very similar leaf death rates. Bouteloua gracilis also
had senescence rates which were very similar in both 1978
78
and 1979. Within each year, senescence rates varied less
among the treatments than did the emergence rates.
Difference Between the Tvo Rates. The net result of
the actions of emergence and senescence may be seen in the
”difference" column of tables 15 and 16. Agropyron smithii
wet treatments in all years had a net increase in leaf
numbers over the start-of-season value (Section III). It
can be seen from table 15 that this increase was due to a
higher mean emergence rate. In 1977, the first year of
irrigation treatments and a time when culm density was
still increasing rapidly in the wet plot (Section I), the
difference between emergence and senescence was larger
than in any other year.
For other treatments in the Agropyron smithii plots,
with one minor exception, mean senescence rates were
higher than emergence rates. The plants lost leaves more
rapidly than they produced them. This is confirmed by the
net intraseasonal decline in mean leaf numbers per culm
shown in Section III.
Bouteloua gracilis plants lost more leaves in the
course of the summer than they gained for both years
studied and in all treatments.
79
Seasonal Influence on Leaf Emergence
and Senescence Rates
Although there was a large weekly fluctuation in both
emergence and senescence rates, preliminary graphical
analysis (not shown) indicated a general seasonal decline
in both rates for both species. In order to document this
trend, rates from all treatments were grouped by month and
the monthly rates were compared. In order to prevent
complication by water stress effects, comparison of mean
monthly emergence rates (Table 17) was made using only
those observations of minimally stressed plants which had
water potentials between 0 and -0.49 MPa.
Table 17. Mean monthly leaf emergence rates for Agropyfon
smithii (1977-1979) and Bouteloua gracilis
(1978-1979). Observations are of plants with
water potentials of 0-0.49 -MPa. Means within
each row with different letters indicate
significant differences with p< .05 .
Species
Agropyron
Bouteloua
June
51.8 A
35.1 A
July
30.8 B
9.8 B
August
19.6 B
17.0 B
Both species had the highest rate of leaf emergence
in June . The rate dropped significantly in July and rates
in July and August could not be differentiated.
Leaf senescence rates for both species were also
grouped by month in a manner analogous to the emergence
rates (Table 18). These means did not seem to have come
80
from different populations so no significant effect of
seasonal trend could be shown.
Table 18. Mean monthly leaf senescence rates for Agropyron
smithii (1977-1979) and Bouteloua gracilis
(1978-1979). Observations are of plants with
water potentials of 0-0.49 -MPa. Differences
between means within rows did not have
significance levels with p > .05 in all
comparisons.
Species
Agropyron
Bouteloua
June
32.4
33.4
July
40.4
35.4
August
22.8
26.3
Effect of Water Potential on Emergence
and Senescence Rates
Graphical Analysis of Emergence. Irrigation
treatments were shown earlier to affect emergence rates
(see above). For this reason , an examination of the
specific correlation of plant water stress and leaf
dynamics was desirable.
Figures 5 and 6 illustrate the
relationship between leaf emergence and plant water
potential for Agropyron smithii and Bouteloua gracilis.
Observations have been grouped by month to avoid
confounding effects of water and season on leaf dynamics.
For both species, leaf emergence never took place at
a high rate when water stress was high. When water stress
was low (0-1.0 MPa) leaf emergence rates showed a broad
range of values (Figure 5 and 6).
Ld
90*
Ld
70-
AUGUST
JUNE
6 -i
-i -4 -A -6
4- (MPa)
Figure
5. Agropyron smithii leaf emergence rate vs. water p o t e n t i a l .Data from
1977 - 1979 are pooled . Emergence rate is calculated as the number
of leaves born in one week per 100 culms. Water potential applies
to the period during which emergence was taking place.
Ul
H-
<
QL
Ul
U
Z
Ul
O
JUNE
AUGUST
QL
UJ
Ul
Lu
<
LU
* (MRa)
Figure 6.
* (MPa)
* (MPa)
Bouteloua gracilis leaf emergence rate vs. water potential. Data from
1978 - 1979 are pooled. Emergence rate is calculated as the number
of leaves born in one week per 100 c u l m s . Water potential applies
to the period during which emergence was taking place.
I O O i --- Q----------------------------------------- I O O l -w-------------------------------------------
100
oz 80
w
o
C
"
Z
c
C
LU
O
V)
U
Z
LU
W
Lu
<
LU
" t wCe
30 -
0
0J
50c
c
W6 %
tfi
6
C
* (MPa)
Fieure
3020-
o-
AUGUST
60-
B
S6
B
70-
JULY
Q
C
50e
•C c
C
40-
40-
10-
W
wc
C
80-
60-
20 _ mW
10
8070-
.J U N E
6
C
40
B
90-
—I—
O
o>
<
CW
30-
#
wf
C
C
20-
W
Bc
C
10-
C
W6
*
* (MPa)
CO
6W
C
6
Bw O - I CBf
I
6
I
B
C
I
I
I
I
* (MPa)
7. Aeropyron smithii leaf senescence rate vs. water potential. Data from
1977 - 1979 are pooled. Senescence rate is calculated as the number
of leaves which died in one week per 100 culms. Water potential
applies to the period during which senescence was taking place.
JUNE
AUGUST
Lu 40-
OO
4N
^ (MPa)
Figure 8.
^ (MPa)
* (MPa)
Bouteloua gracilis leaf senescence rate vs. water potential. Data
from 1978 - 1979 are pooled . Senescence rate is calculated as the
number of leaves which died in one week per 100 c u l m s . Water potential
applies to the period during which senescence was taking place.
f
85
Graphical Analysis of Senescence. Grouping senescence
observations by month and pooling all treatments , figures
7 and 8 show there was little correlation of water stress
with senescence rate for either Agropyron smithii or
Bouteloua gracilis. Leaf senescence rates showed much less
correlation with water stress than did emergence. While
high senescence rates were expected when plants were
highly water stressed, this was not found consistently.
High rates of senescence took place at all levels of water
stress for both species.
Statistical Analysis. To lessen the confounding
-effect of season on the relationship between water
potential and leaf dynamic rates, only data from June were'
used for statistical analysis. Tables 19 and 20 compare
mean emergence and senescence rates at for Agropyron
smithii and Bouteloua gracilis at three levels of water
stress.
Table 19. Mean June leaf emergence and senescence rates
for three levels of water stress in Agropyron
smithii. Rates are calculated as leaves emerged
(or senesced) per IOO culms in a one week
period. Different letters within the same column
indicate means with a probability of belonging
to the same population at p < .05.
Water Potential
(MPa)
0 to -.49
-0.5 to -1.99
< - 2.0
Emergence rate
51.8 A
19.5 B
3.2 B
Senescence rate
27.7 A
26.4 A
27.2 A
J
86
Table 20. Mean June leaf emergence and senescence rates
for three levels of water stress in Bouteloua
gracilis. Rates are calculated as leaves
emerged (or senesced) per 100 culms in a one
week period. Different letters within the
same column indicate means with a probability
of belonging to the same population at p < .05.
Water Potential
(MPa)
0 to -.19
-0.2 to -.49
-0.5 to -.99
Emergence rate
Senescence rate
26.7 A
39.4 A
32.0 A
22.6 A
38.9 B
32.1 AB
A comparison of means confirms conclusions of the
graphical analysis for Agropyron smithii. Although the
emergence -rates were variable at high water potentials (0
to -0.49 MPa) the mean was significantly higher for this
class. Between lower water potential classes (-0.5 to
-1.99 MPa and below -2.0 MPa) effect of water stress on
emergence was not statistically differentiable. Senescence
rates did not differ statistically among the three levels
of water stress.
The Bouteloua gracilis study had a narrower range of
water potential values to be matched.with observed
emergence rates than Agropyron smithii study had. With the
values available,
no effect of water stress on leaf
emergence could be shown statistically. Senescence rates
were significantly higher in the -0.2 to -.49 MPa range of
water potentials than the more stressed -0.5 to -.99 MPa
“ range. No explanation for this unexpected observation is
offered.
87
Leaf Juvenile Period
The juvenile period of a leaf was defined as the time
between a leaf's emergence and the time when the leaf
ceased to grow longer . Because this period is expressed in
days and measurements were taken at approximately weekly
intervals the reader should recognize the possibly
overstated precision of these figures.
At any given moment there were rarely more than two
or three immature leaves on a single culm. Leaves tended
to be produced sequentially; each leaf at least nearing
attainment of it's ultimate size -before a new leaf
emerged.
The hypothesis that supplemental water would decrease
the time required for leaf elongation is supported by the
results of 1977'(Table 21). The control plants required
the longest time for leaf development, the 6 mm plants
took less time than those in the the control plots, and
the wet treatment plants took the least time of all.
Table 21. Mean juvenile period (days) for leaf cohorts of
Agropyron smithii.
Emergence date
7 June 1977
15 June 1978
5 June 1979
______________ Treatment
6mm
12mm
Control
44
56
21
17 "
18
27
27
16
—
Wet
17
19
19
88
Data from the year 1978, which had above normal
rainfall in the spring , are also consistent this
hypothesis. Plants in all treatments took roughly the same
time for their leaves to mature, 18 days. This interval is
also roughly the same as that for the 1977 wet treatment
plants , 17 days.
Data from 1979 tends to refute the development rate
hypothesis. With weather conditions in this year dryer
than normal, control plants had the shortest mean juvenile
period observed. Plants in the 6 mm and 12 mm plots had
leaves' with the longest mean juvenile period,, while plants
in the wet treatment had an intermediate maturation time.
I have no explanation for this discrepancy, and on the
basis of the conflicting results, the hypothesis that
supplemental water will decrease leaf juvenile intervals
cannot be accepted.
Leaf Adult Interval
The interval between the moment when a leaf reaches
its maximum length and the moment at which the leaf has
senesced was defined as the leaf's "adult phase". The
hypothesis was advanced that plants which received
supplemental water would have leaves which had longer
adult phases. For example, leaves of non-water-stressed
plants might persist until the first frosts while plants
with stress might shed leaves-to conserve water. This wet
89
plot culm behavior would be possible only as long as culm
density was low enough to prevent lower leaves from being
shaded out. If the culm density became higher, leaf adult
phase would be shortened due to shading of the leaves.
To test this hypothesis on Agropyron smithii, it was
necessary to use leaves of the same rank order, as was
done in Section III for maximum leaf elongation analysis.
Robson (1973) demonstrated a possible relationship between
leaf rank and leaf adult duration in the perennial
ryegrass Lolium perenne. If leaves of different rank were
grouped the factor of leaf rank would confound analysis of
adult period. Unfortunately this restriction reduced the
number of leaves available for consideration to such a
small sample size that analysis was deemed unfeasible.
I
90
DISCUSSION
Because both Agropyron smithii and Bouteloua gracilis
produce new leaves only at the apical meristem, leaf
emergence was calculated on the basis of a standard number
of culms producing a variable number of leaves in a
standard interval. One might calculate leaf senescence
rates differently, reasoning that those culms (and
treatments) with more leaves would also have more leaves
which were at risk of death and thus be biased towards
higher senescence rates. This would be inappropriate for
these grass species since virtually without exception,
leaves died in order - from the bottom up. This allows
calculation of senescence rates to be exactly analogous to
that for emergence.
When comparing the mean weekly rates of leaf dynamics
between years, interpretation of leaf emergence and
senescence dynamics is complicated by both culm density
variation and seasonal changes. First, because of the
higher culm density which developed as the irrigation
progressed (Section I), the high rate of leaf production
shown by plants in the wet plot in 1977 could not be
maintained in 1978 even though conditions were more moist.
Second, due to different seasonal rainfall levels, plants
91
in unwatered plots are expected to perform differently in
wet and dry years .
For both species, leaf emergence rates were
positively correlated with low plant water potential. This
was expected because water stress inhibits many of the
functions required by plant growth (Hsio 1973). The
absence of even moderate emergence rates when water stress
was high shows that water is a true limiting factor whose
absence limits the ability to produce new leaves. The wide
range of emergence rates which accompanied low water
stress indicates that water is not the only limiting
factor.
A decline in emergence rate with season was expected
on the basis of plant energetics. Energy investment in
leaf production at the end of the year makes little
evolutionary sense. Conditions then are less favorable for
photosynthesis due to xeric conditions and the increased
possibility of tissue killing frost minimizes the
potential for return on the energy investment.
That leaf senescence was not correlated with season
can also be explained in terms of plant energetics. The
energy required to produce a leaf is undoubtedly greater
than the energy required to let the leaf wither. Once the
investment of energy required for leaf synthesis is
expended, it is reasonable that a. plant might allow the
leaf to remain green as long as the leaf can produce a net
92
energetic surplus. Consideration of the anatomy of grass
plant's supports this conjecture in that grass leaves lack
the abscission layer used by many angiosperms to remove
leaves from the main structure of the plant.
93
CONCLUSIONS
Under non-water-stressed conditions,
leaf emergence
was highly correlated with season for both Agropyron
smithii and Bouteloua gracilis. Emergence rates in June
were higher than those found in the rest of the season.
Leaf senescence rates were not correlated with season.
Emergence rates were correlated with plant water
potential for Agropyron smithii, but were not correlated
for Bouteloua gracilis. The wide range of emergence rates
found when the plants had plenty of water suggests that
several factors other than water stress influence this
process. When water stress was high emergence rates were
always low, as expected if water were a limiting factor.
Leaf senescence rates were no higher under dry than wet
conditions.
The time required for a newly emerged leaf to
elongate fully was shortened by addition of supplemental
water in 1977 but was not in 1978 and 1979. No explanation
for the discrepancy is offered, but this negates the
hypothesis of accelerated leaf development.
Data from this study did not permit comparison of the
duration of the mature phase of leaf life history among
irrigation treatments.
94
REFERENCES CITED
95
'I
REFERENCES CITED
Antonivics, J., D .A . Levin
1980. The ecological and
genetic consequences of density-dependant regulation
in plants. Ann. Rev. Ecol. Syst. 11:411-52.
Barge, B.L., B . Kochtubajda, M . English, J. Renick, and
R . Humphries
1986. Potential for weather
modification. Atmospheric Sciences Report 86-1.
Alberta Research Council, Natural Resources Division.
Bazzaz , F .A . and J .L . Harper
1977. Demographic analysis
of the growth of Linum usitatissimum. New
Phytologist 78:193-208.
Bradshaw, A .D . 1965. Evolutionary significance of
phenotypic plasticity in plants. A d v . in Genetics
13:115-155.
Harper, John L . 1977. Population Biology of Plants.
Academic Press pp. 892
Hsiao, T.C.
1973. Plant response to water stress. Ann.
Rev. Plant Physiol. 24:519-570.
Kays, S . and J .L . Harper
1974. The regulation of plant
and tiller density in a grass sward. Journal of
Ecology. 62:97-105.
Kramer, P .J . 1969 . Plant and Soil Water Relationships: A
Modern Synthesis. McGraw-Hill pp. 482.
Newbauer, J . 1985. Growth respons.e of Agropyron smithii
to increased summer water availability. Masters
Thesis, Montana State University, Bozeman, MT, 1985,
p p . 43 .
96
Perry , D . 19 76. The effects of weather modification on
northern Great Plains grasslands: preliminary
assessment. J n l . Range M g t . 29:272-278.
Ritchie, G.A. and T.M. Hinkley
1978. The pressure
chamber as an instrument for ecological research.
In: Advances in Field Research. (J. Gragg, ed.)
Academic Press, "pp. 165-254.
Robson, M.J.
1973. The growth and development of
simulated swards of perennial ryegrass. I. Leaf
growth and dry' weight change as related to the
ceiling yield of a seedling sward. Ann. Botany
37:487-500.
Silvertown, J.W.
1982. Introduction to Plant Population
Ecology. Longman, pp. 209
Simpson, G.M.
p p . 324.
1981. Water Stress on Plants. Praeger,
Snedecor, G.W. and G.W. Cochran
1980. Statistical
Methods. Seventh edition. Iowa State U n i v . Press.,
pp . 507 .
Taylor, S., D. Evans, and W . Kemper
1961. Evaluating soil
water. Ag. Expt. Station Bull. 426. Utah State Univ.,
Logan, p p . 64.
»
Weaver, T.W., J. Birkby, J . Welker, J. Newbauer
1981.
Short term responses of Agropyron smithii vegetation
to six water regimes. In: State of MT Activities in
the High Plains Cooperative Program: 1981-1983, final
report (Kudsem and M o y , eds.)
p p . 103-109.
Weaver, T.W.
1983. Response.of differentially enriched
grasslands to cessation of irrigation- dry years
following wet. In:- State of MT Activities in the
High Plains Cooperative Program: 1981-1983, final
report (Holman and Gerhard, eds.)
pp. 95-101.
97
Webster, B .D . 1973. Anatomical and histochemical changes
in leaf abscission. In: Shedding of Plant Parts.
(T.T. Kozlowski, ed.) Academic Press. p p . 560.
White, J . 1979. The plant as a metapopulation. A n n .
Rev . Ecol. Sys t . 10:109-145.
Whittaker, R.H.
1975. Communities and Ecosystems.
Second edition. MacMillan Publishing Company,
p p . 385 .
APPENDICES
99
APPENDIX A
Field Observations
The following descriptions of methods pertain to
details of field observations.
Early in the field season, representative culms (20
for Agropyron smithii and 30 for Bouteloua gracilis) were
selected for observation and marked. Due to the clonal
nature of these plants it is impossible to ascertain the
genetic diversity represented by this sample. As these
culms were all roughly the same size and possessed of
roughly the same number of leaves early in the season, I
have assumed them to be of the same age. Thus all leaf
measurement samples are based on an early season cohort of
culms. This same cohort was followed for the rest of the
season .
Plants were observed at approximately one week
intervals during the years 1977, 1978, and 1979 for
Agropyron smithii, 1978 and 1979 for Bouteloua gracilis.
Early in the season,' individual culms were selected as
being representative of the stand, tags were inserted at
their base and their development was followed from roughly
mid May until 30 August. These culms were spaced down the
middle of the treatment row and spaced approximately 75 cm
apart.
Leaves on each culm were evaluated for physical
condition and length. Each leaf was also assigned a number
designating the leaf’s relative position on the culm, with
"#I" being the leaf closest to the ground and counted
acropetally in ascending order. In this way, subsequent
observations could be made on the same leaf for the entire
season. Leaf length was measured with a ruler and was
defined as the distance from the ligule of the lower leaf
to the outstretched tip of the measured leaf, held along
the culm axis. Culm height was measured by lifting the
grass away from the ground and measuring the distance
between the ground and the tip of the most outstretched
leaf .
Collection of this field data was done principally by
John Newbauer, who was assisted by various personnel on
the weather modification project.
Field sheets were transcribed to a computer readable
format with alpha numeric codes used to designate leaf
condition. (Appendix B )
100
Approximately three or four weeks into the field
season, specified leaves (usually leaf #3 or #4) of these
plants were trimmed at an angle in order to facilitate
identification of the other leaves based on their relative
position along the culm. These leaves have been marked
with an "X" in the third column of their four character
condition code. The first appearance of the "X" designates
the date at which the cut was made. Once cut, the leaf
continued to be marked with the "Xff to indicate that the
leaf was altered.
Each data file represented a year's observations for
a species-treatment. After entry was complete, data were
proofread for transcription errors by reading from the
original field sheets and checking them against the
computer print out.
Further data checking was performed by sorting the
data by plant number and running a "transition error"
checking program (CHKDAT) on the sorted data. This program
read evaluations of leaf conditions for paired dates. If
the data revealed that on successive dates the leaf
changed condition in a manner that was deemed not to be
obviously part of the naturally occurring maturation
process of leaf development, the pair of observations was
written to a file for later evaluation and correction.
The corrections were made on a case by case basis,
using the following assumptions.
1.
) Leaves were not able to shrink except as a result
of damage and lost tissue.
2. ) A leaf once lost must remain lost.
3. ) A leaf once turned brown was not able to
regenerate.
4.
) In a series of observations, if a leaf turns
brown, then appears to turn green again, the observation
with the brown designation is the suspect one.
5.
) If a leaf's length measurement was deleted for
some reason, its condition designation could be allowed to
stand. The inverse was not allowed to occur.
Certain plants were replaced during the course of the
field season, usually during the first three observation
dates. In cases where notes were provided stating that
marking tags were missing (eg. rodent damage to tags) past
information on these plants was discarded and the newly
marked plants were put in their place with the assumption
that the new plants were alive early in the season, but
with unknown condition. The newly marked plants have a "B"
placed after the plant number.
Certain plants died early in the field season and
were replaced with new plants. The dead plants were kept
in the sample population, designated "DEAD", and an "A"
was placed after the plant number. The new plant, as in
the case with the rodent damaged tags, was given the
101
suffix of "B" after its number and assumed to be alive
early in the season but with unknown condition.
Some plants were replaced for reasons that remain
unclear. If the plant's previous condition was healthy, it
was assumed not to have died, and was simply removed from
the sample. Its replacement was given the suffix "B" and
assumed to be alive on the date of the season's first
observation. If the plant did not appear healthy, but was
replaced without explicitly being called dead, death was
presumed. If several leaves were marked as green but
damaged, death was assumed to occur one observation
following the introduction of the new, remarked plant. The
dead plant was given the suffix "A" and the replacement
"B" as in the previous situations.
102
APPENDIX B
Coding of Observations and Data Files
The following codes were used on the field sheets. In
cases where the code was modified during computer data
entry, the new code is indicated with an "=" symbol.
A- developing leaf in the highest position on culm.
AB- mature leaf in the highest position on culm
AC- developing leaf in highest position with damage
in excess of 2 mm.
ACL- developing leaf in highest position with a
damaged tip which has been lost
FH=F- Flag leaf
CF- damaged flag leaf.
ABC- mature leaf in highest position that was
damaged.
B- mature leaf.
CLF- flag leaf with a damaged tip that has been lost.
C- leaf with more than 2 mm damage.
CL- leaf that has Lost more than 2 mm from tip.
D- yellow leaf. Red=D- yellow leaf.
E- leaf more than 75% brown.
F=G- grey leaf.
L- leaf that has become lost.
H- seed head.
Leaves were assigned the following character codes to
describe their condition. N,o other codes are possible in
the data set. To verify that this was in fact the case
program CHKSYMB was run on all the data files to verify
that no other symbols were used. For some cases where the
height of seed heads were available from the Bouteloua
files the height was recorded in mm directly following the
"H" "symbol.
"A
"- green developing leaf in the highest position
on the culm.
"F
"- flag leaf , always in the highest position on
culm
"CF "- damaged flag leaf
"EF "- flag leaf over 75% brown
"CLF "- flag leaf with' more than 2 mm lost from tip
"AB "- green mature leaf in highest position on culm
"ACL "- green developing leaf in highest position
with more than 2 mm lost from tip
103
"ABC
green mature leaf in highest position with
more than 2 mm damage
"ABCL"- green mature leaf in highest position with
more than 2 mm damage on tip that has been lost .
11B
green mature leaf
"CL "- leaf that has lost more than 2 mm from tip
"C
"- leaf with damage done to it
"G
"- gray leaf
"E
"- leaf
more than 75% brown
"D
"- yellow or redleaf
"L
"- lost leaf
"B X "- green mature leaf that has been clipped for
marking
"C X "- damaged leaf that has been clipped for
marking
"CLX "- leaf with damaged tip that has been clipped
"D X "- yellow leaf that has been clipped
"E X "- brown leaf that has been clipped ■
"L X "- lost leaf that has been clipped
"H
"- seed head
"?
"- leaf with unknown condition, might not even
exist, missing data
"? X "- leaf with undetermined condition that had
been clipped in past
The data were entered from field notes onto computer
disks in 20 data files.
1977
ASCN 77.DAT
AS6M77.DAT
ASWT77.DAT
1978
ASCN78.DAT
AS6M78.DAT
AS 1278.DAT
ASWT78.DAT
1979
ASCN79.DAT
AS6M79.DAT
AS 1279.DAT
ASWT79.DAT
ASFC79.DAT
ASFW79.DAT
BGCN78.DAT
BG6M78.DAT
BGWT78.DAT
BGCN79.DAT
BG6M79.DAT
BGWT79.DAT
1981
ASCN81.DAT
MONTANA STATE UNIVERSITY LIBRARIES